JP2004032463A - Method for dispersively speech amplifying to localize sound image by following to speaker movement and dispersively speech amplifying system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a method for dispersively speech amplifying to localize a sound image by following speaker movement which can change the direction of sound image localization by matching the direction of a moving sound source, and to provide a dispersively speed amplifying system. <P>SOLUTION: The method for dispersively speech amplifying to localize the sound image by following the speaker movement includes steps of dispersively disposing a plurality of speakers Sj having substantially equal acoustic characteristics used for a loudspeaker of a sound of a sound source O downward at a predetermined position above a sound receiving surface Ph, setting a site on the surface Ph vertically under the speakers Sj to a sound receiving point Pi, detecting the position of the source O moving in a sound field in which the speakers Sj are dispersively disposed, selecting the speaker nearest to the detected position of the source O of the speakers Sj as a main speaker So and sequentially ordering the residual speakers to the speakers Sgx in raising order of a distance from the speaker So, generating a loudspeaker sound of the source O from the speaker So at time zero without a time delay via a required input power, and time delaying the speakers Sgx sequentially in raising order and generating via the required input power. <P>COPYRIGHT: (C)2004,JPO

Description

【００２５】
図１では、音圧周波数特性、指向係数等の音響特性が同一のスピーカＳｊのみを使用している。図３（ｂ）のグラフは、正面軸に対し音圧レベルＬが−６ｄＢ減衰する開き角度（指向角）の指向周波数特性を示す。例えば図３（ｂ）の指向周波数特性のスピーカを本発明の分散配置スピーカＳｊとして選択した場合、１ｋＨｚの指向角が８５〜９０度程度であるから、その指向角とカバーエリアの重畳率とを考慮して各スピーカＳｊの受音面Ｐｈからの配置高さを設計することができる。図１のスピーカ配置は、図３に示す指向周波数特性のスピーカを分散配置した場合の一例である。
【００２６】
ただし、本発明で用いるスピーカＳｊは必ずしも同一種類のものに限定されない。例えば１ｋＨｚを中心とする１オクターブバンド、好ましくは５００Ｈｚ〜２ｋＨｚのバンドの音質の違いが少なく、実質的に同一音響特性であるときは、種類の異なるスピーカを用いることができる。
【００２７】
分散配置スピーカＳｊの相互間隔は、受音点に対する音源Ｏ方向の２つのスピーカＳｊの伝送特性が類似するように定める。後述するように本発明では、音源Ｏに最も近いスピーカＳｊを先行音スピーカとし、音源Ｏから受音点（聴者２）へ向かう方向の周辺のスピーカＳｊを後続音スピーカとして音像を定位する。配置高さとの関係を考慮しつつ分散配置スピーカＳｊの間隔を適当に設計することにより、受音面Ｐｈ上の任意の受音点において、音源方向の先行音スピーカと後続音スピーカの両耳に対する伝送特性の差を音像定位が可能な許容範囲内とすることができる。
【００２８】
さらに、上述したスピーカＳｊの配置において、受音点から見た音源Ｏの方向と先行音スピーカの方向とのずれを図４に示す許容範囲内に収める。例えば図１では、スピーカＳ５の下方の聴者２から見て、話者Ｏが他の何れのスピーカＳｊの下方にいる場合でも、話者Ｏの方向に対し話者上方のスピーカＳｊが図４の許容範囲内に配置されているので、話者上方のスピーカを先行音スピーカとして音像定位を得ることができる。すなわち本発明では、複数のスピーカＳｊを上記カバーエリアの重畳を得る高さと、類似の伝送特性を得る相互間隔と、先行音スピーカの配置許容範囲とを満たす所定位置に分散配置する。
【００２９】
分散配置した各スピーカＳｊには、信号伝送装置１０により音源Ｏからの音響信号が入力される。図１の信号伝送装置１０は、マイクロホン３とミキサー１１と信号伝送路とを有する。マイクロホン３は音場内の話者や演奏者その他の音源Ｏの近傍に配置して直接音を収音する。収音した直接音は、マイクアンプ（図示せず）で増幅したのち、ミキサー１１により音場全体の音量コントロールが施され、後述の信号調整装置２０へ送られる。ミキサー１１でのマイク音量コントロールは、音源Ｏの音量を考慮して、オペレータが手動でコントロールできる。ただし、本発明の適用範囲は音場内の音源Ｏの音の拡声に限定されず、例えば予め録音した音源から拡声により音像を定位させる場合等にも適用できる。
【００３０】
本発明では、信号伝送装置１０と各スピーカＳｊとの間に、各スピーカＳｊの音響遅延と音圧レベルを調整する信号調整装置（イフェクタ）２０を挿入する。信号調整装置２０の一例は、図１３に示すように、コンピュータ３０（図１４参照）からの指示信号の入力に応じて、スピーカＳｊ毎に音響信号の発音時刻と入力パワーとを調整する音響遅延回路２１及び音圧レベル制御回路（音量制御回路）２２を有するものである。各信号調整装置２０はコンピュータ３０に接続され、コンピュータ３０より各信号調整装置２０の音響遅延と入力パワーの調整とが一括して制御される。
【００３１】
図５は、音場内の音源Ｏの音の拡声により音像定位を得る場合におけるコンピュータ３０による信号調整装置２０の制御の流れ図の一例を示す。以下、図５の流れ図を参照して本発明の拡声方法を説明する。先ずステップ６０１において音場内の音源Ｏの位置を検出する。図１では、マイクロホン位置検出装置３７の検出信号に基づき音源Ｏの位置を検出している。
【００３２】
すなわち、マイクロホン３に付設した赤外線発信タグ１０１からはタグ識別番号が送出される。このタグ識別番号は、マイクロホン３が位置しているカバーエリアのタグ読み取り装置１０３によって検出される。タグ読み取り装置１０３は、このタグ識別番号と、自身に割り当てたリーダ識別番号とをコンピュータ３０へ送出する。コンピュータ３０は、音源位置同定手段１１０によってこれらの識別番号からマイクロホン３の位置を検出する。つまり、マイクロホン３の位置は、マイクロホン位置検出装置３７によって自動検出される。
【００３３】
次にステップ６０２において、スピーカ選択手段３１により音源Ｏに最も近いスピーカＳ１を主スピーカＳｏとして選択する。スピーカ選択手段３１の一例は、音源Ｏと各スピーカＳｊとの間の距離を計算して距離最小のスピーカＳ１を選択するコンピュータ内蔵のプログラムである。
【００３４】
ステップ６０３では、例えばコンピュータ３０内蔵のプログラムであるスピーカ順位付け手段３２により、主スピーカＳｏ以外の分散配置スピーカＳｊを主スピーカＳｏからの距離の昇順に順位付けする。本明細書において、順位付けされたスピーカＳｊを周辺スピーカＳｇｘ（ｘ＝１、２、．．．．．．、（ｎ−１））という。例えば図１では、分散配置のスピーカＳ２を音源Ｏに一番近い周辺スピーカＳｇ１とし、スピーカＳ３を４番目に近い周辺スピーカＳｇ４とし、スピーカＳ４を８番目に近い周辺スピーカＳｇ８とし、スピーカＳ５を１５番目に近い周辺スピーカＳｇ１５として順位付けしている。
【００３５】
ステップ６０１〜６０３に示すように、本発明では主スピーカＳｏの位置を固定せず、音源Ｏに最も近い分散スピーカを主スピーカＳｏとし、主スピーカＳｏからの距離に応じて周辺スピーカＳｇｘを順位付けするので、音源Ｏが移動した場合でも移動に応じて主スピーカＳｏの位置及び周辺スピーカＳｇｘの順位を変化させ、受音面Ｐｈ上における音像定位を変化させる。すなわち、移動する音源Ｏに追従して音像定位することができる。
【００３６】
ステップ６０４において、例えばコンピュータ３０内蔵のプログラムである主スピーカ音指示手段３３により、主スピーカＳｏの発音時刻Ｔｏ及び入力パワーＷｏを指示する。例えば主スピーカＳｏで信号伝送装置１０からの音響信号を時間遅延なく発音し、その発音時刻Ｔｏを時刻零（ｍｓ）とおく。ただし主スピーカＳｏの発音時刻Ｔｏに時間遅延を設けても、本発明の音像定位を得ることが可能である。また主スピーカＳｏの入力パワーＷｏは、マイクロホン３における主スピーカＳｏからの拡声音到達音圧レベルが、話者Ｏから肉声の到達音音圧レベル（１ｍ離れた地点の音圧レベル）に比べてハウリング防止に要するレベル、例えば２ｄＢだけ低いレベルとなるように決定する。ただし、ハウリングの発生がない場合は主スピーカＳｏの入力パワーＷｏを所要レベルとすることができる。主スピーカＳｏの入力パワーＷｏの決定に際し、マイクロホン３に対応するミキサー１１のボリュームを適宜調整でき、その際のアンプ１２の減衰器（ａｔｔｅｎｕａｔｏｒ）の減衰量（以下、アッテネータ位置ということがある。）を０（ｄＢ）とおくことができる。
【００３７】
ステップ６０５以降の周辺スピーカ音指示手段３４による各周辺スピーカＳｇｘの発音時刻及び入力パワーの決定方法については、説明簡単化のため、話者Ｏと聴者２との間に５つのスピーカＳ１〜Ｓ５を配置した図６を参照して説明する。図６は、話者Ｏに最も近いスピーカＳ１を主スピーカＳｏとし、スピーカＳ２〜Ｓ５を周辺スピーカＳｇ１〜Ｓｇ４として順位付けし、周辺スピーカＳｇ４の下方に聴者２がいる場合である。ステップ６０５では、先ず各スピーカＳ１〜Ｓ５の垂直下方の受音面Ｐｈ上に受音点Ｐ１〜Ｐ５を定める。図６において、受音点Ｐ１は主スピーカＳｏの垂直下方の受音点Ｐｏ、受音点Ｐ２〜Ｐ５は周辺スピーカＳｇｘ（Ｓｇ１〜Ｓｇ４）の垂直下方の受音点Ｐｇｘ（Ｐｇ１〜Ｐｇ４）である。
【００３８】
主スピーカＳｏからの拡声音が周辺スピーカＳｇｘ下方の受音点Ｐｇｘに到達するに要する時間は、主スピーカＳｏと受音点Ｐｇｘとの間の距離をＲｏｘとすれば、下記（１０）式として算出できる。（１０）式において符号Ｃは音速（ｍ／ｓ）を表す。通常は音速Ｃとして常温・常湿（２３℃、相対湿度６０％）の音速を用いる。また、受音点Ｐｇｘに到達した主スピーカＳｏからの拡声音の音圧レベルＬｏｘは、式（４）を参照して下記（１１）式として表すことができる。（１１）式において、Ｗｏは主スピーカＳｏの入力パワー、Ｄｏは主スピーカＳｏの指向係数を示す。
【００３９】
また、周辺スピーカＳｇｘからその下方の受音点Ｐｇｘへの拡声音の到達に要する時間及び到達音圧レベルＬｘｘも、周辺スピーカＳｇｘと受音点Ｐｇｘとの間の距離をＲｘｘとすれば、主スピーカＳｏの場合と同様にして（１２）式、（１３）式として表わすことができる。（１３）式において、Ｗｘは周辺スピーカＳｇｘの入力パワー、Ｄｘは周辺スピーカＳｇｘの指向係数を示す。なお図示例では、主スピーカＳｏと周辺スピーカＳｇｘとは同一の音響特性であるから、（１１）式と（１３）式のＡ（ｆ）は同じである。
【００４０】
各受音点Ｐｇｘには、主スピーカ及び複数の周辺スピーカからの拡声音が到達し得る。複数の同一特性Ａ（ｆ）のスピーカＳｊ（ｊ＝１、２、３、．．．．．．）からの拡声音が到達する受音点Ｐｇｘでの合成音圧レベル（以下、聴取音圧レベルという。）は、各スピーカＳｊからの到達音圧の自乗音圧Ｉの総和に基づき、下記（１４）式で表すことができる。
【００４１】
【数２】

【００４２】
ステップ６０５〜６１２では、各周辺スピーカＳｇｘの順位の昇順に従い、各周辺スピーカＳｇｘの発音時刻及び入力パワーを決定する。先ず図６の周辺スピーカＳｇ１について、ステップ６０５で主スピーカＳｏから周辺スピーカＳｇ１の下方の受音点Ｐｇ１までの距離Ｒｏ１と、周辺スピーカＳｇ１から受音点Ｐｇ１までの距離Ｒ１１を算出する。周辺スピーカＳｇ１の場合は、下位順位の周辺スピーカＳｇｘ、すなわち周辺スピーカＳｇ１よりも更に主スピーカＳｏに近い周辺スピーカＳｇｘは存在しない。ステップ６０６では、主スピーカＳｏから受音点Ｐｇ１までの距離Ｒｏ１の（１０）式ヘの代入結果と主スピーカＳｏの発音時刻Ｔｏ＝０（ｍｓ）とから、受音点Ｐｇ１における主スピーカＳｏからの拡声音到達時刻Ｔｏ１を算出する。また、主スピーカＳｏから受音点Ｐｇ１までの距離Ｒｏ１と主スピーカＳｏの入力パワーＷｏ及び指向係数Ｄｏとの（１１）式ヘの代入結果から、受音点Ｐｇ１における主スピーカＳｏからの拡声音到達音圧レベルＬｏ１を算出する。ただし主スピーカＳｏの音圧周波数特性Ａ（ｆ）は所定とする。
【００４３】
ステップ６０７では、主スピーカＳｏからの拡声音到達時刻Ｔｏ１に対し先行音効果を与える遅延時間Δｔ１を求める。この遅延時間Δｔ１は、主スピーカＳｏに最も近い距離の周辺スピーカＳｇ１で最短遅延時間Δｔｍｉｎ、例えば３ｍｓとし、主スピーカＳｏから最も遠い距離の周辺スピーカＳｇ４で最長遅延時間Δｔｍａｘ、例えば１０ｍｓとし、他の周辺スピーカＳｇ２、Ｓｇ３では主スピーカＳｏから当該周辺スピーカＳｇ２、Ｓｇ３下方の受音点Ｐｇ２、Ｐｇ３までの距離Ｒｏ２、Ｒｏ３に比例配分した時間として定めることができる。
【００４４】
ただし遅延時間Δｔ１の決定方法は前記距離に応じた比例配分による方法に限定されない。例えば、主スピーカＳｏから周辺スピーカＳｇ２、Ｓｇ３下方の受音点Ｐｇ２、Ｐｇ３までの距離Ｒｏ２、Ｒｏ３の対数（ｌｏｇＲｏｘ）に比例した最短遅延時間と最長遅延時間との間の時間として定めることができる。下記表１は、この関係から定めた各周辺スピーカＳｇ１〜Ｓｇ４の遅延時間Δｔｘを示す。ただし表１では、最短遅延時間Δｔｍｉｎである周辺スピーカＳｇ１の遅延時間Δｔ１を６ｍｓとし、最長遅延時間Δｔｍａｘである周辺スピーカＳｇ４の遅延時間Δｔ４を３０ｍｓとしている。
【００４５】
求めた遅延時間Δｔ１と主スピーカＳｏからの拡声音到達時刻Ｔｏ１との和として、ステップ６０７において、周辺スピーカＳｇ１からの拡声音が受音点Ｐｇ１に到達すべき到達時刻Ｔ１１（＝Ｔｏ１＋Δｔ１）を算出する。拡声音の到達すべき時刻Ｔ１１が分かれば、周辺スピーカＳｇ１から受音点Ｐｇ１までの距離Ｒ１１を（１２）式へ代入することにより、ステップ６０８において拡声音到達時刻Ｔ１１に対応する周辺スピーカＳｇ１の発音時刻を決定することができる。
【００４６】
【表１】

【００４７】
ステップ６０９において、受音点Ｐｇ１に周辺スピーカＳｇ１からの拡声音到達時刻Ｔ１１より先行して到達する下位順位の周辺スピーカＳｇｋ及び主スピーカＳｏからの拡声音到達音圧レベルの総和を求める。この場合は、受音点Ｐｇ１において周辺スピーカＳｇ１からの拡声音到達時刻より先行して到達する拡声音は主スピーカＳｏからの拡声音のみであるから、主スピーカＳｏからの拡声音の到達音圧レベルＬｏ１を先行する拡声音の到達音圧レベルの総和とし、ステップ６１０において主スピーカＳｏからの拡声音到達音圧レベルＬｏ１に対し先行音効果を与える音圧レベル差ΔＬ１を算出する。この音圧レベル差ΔＬ１は、例えば図８に示すように、先行音効果を与えるに必要な先行音から後続音までの遅延時間Δｔと先行音・後続音間の音圧レベル差ΔＬとの関係式（ΔＬ＝ｆ（Δｔ））を予め実験等により定め、周辺スピーカＳｇ１の発音時刻の算出に要する遅延時間Δｔ１を図８の関係式へ代入（ΔＬ１＝ｆ（Δｔ１））することにより定めることができる。
【００４８】
求めた音圧レベル差ΔＬ１と主スピーカＳｏからの拡声音到達音圧レベルＬｏ１（＝１０ｌｏｇＩｏ１）とから、ステップ６１２において周辺スピーカＳｇ１からの拡声音が受音点Ｐｇ１において示すべき音圧レベルＬ１１（＝１０ｌｏｇＩｏ１＋ΔＬ１）を算出し、算出した到達音圧レベルＬ１１と周辺スピーカＳｇ１から受音点Ｐｇ１までの距離Ｒ１１とを（１３）式へ代入することにより、ステップ６１２において周辺スピーカＳｇ１の入力パワーＷ１を決定する。
【００４９】
例えば図８では、周辺スピーカＳｇ１の到達時間の遅延時間Δｔ１を６ｍｓとした場合、周辺スピーカＳｇ１からの到達音圧レベルと主スピーカＳｏからの到達音圧レベルとの差が６ｄＢ以内であれば音像定位が得られることを示す。主スピーカＳｏの受音点Ｐｇ１における到達音圧レベルＬｏ１は、（１１）式に示すように、距離減衰などにより下がっている。主スピーカＳｏの受音点Ｐｇ１における到達音圧レベルＬｏ１を基準とし、受音点Ｐｇ１において周辺スピーカＳｇ１からの音圧レベルが（Ｌｏ１＋６）（ｄＢ）となるように周辺スピーカＳｇ１の相対入力レベルをアンプ１２のアッテネータ位置により決めた結果が表１に示す−３（ｄＢ）である。このように、周辺スピーカＳｇｘの入力パワーがステップ６１１で算出した値となるように、アンプ１２のアッテネータ位置を調整し又は補正することができる。
【００５０】
ステップ６１２で周辺スピーカＳｇ１の発音時刻と入力パワーとを決定したのち、ステップ６１３ヘ進み、全ての周辺スピーカについて発音時刻と入力パワーの算出が終了したか否かを判断する。この場合は周辺スピーカＳｇ２〜Ｓｇ４の処理が残っているのでステップ６０５へ戻り、ステップ６０５〜６１２を繰り返すことにより、周辺スピーカＳｇ２について発音時刻及び入力パワーを決定する。周辺スピーカＳｇ２の場合は下位順位の周辺スピーカＳｇ１が存在するので、ステップ６０５において主スピーカＳｏから受音点Ｐｇ２までの距離Ｒｏ２、周辺スピーカＳｇ２から受音点Ｐｇ２までの距離Ｒ２２の算出に加えて、下位順位の周辺スピーカＳｇ１から受音点Ｐｇ２までの距離Ｒ１２を算出する。またステップ６０６において、下位順位の周辺スピーカＳｇ１からの拡声音が受音点Ｐｇ２に到達する時刻、及び拡声音到達音圧レベルＬ１２を（１２）式及び（１３）式に基づき算出する。
【００５１】
ステップ６０７において主スピーカＳｏからの拡声音到達時刻Ｔｏ２に対し先行音効果を与える遅延時間Δｔ２を求めるが、この算出方法は周辺スピーカＳｇ１について上述した方法と同様である。この遅延時間Δｔ２と主スピーカＳｏからの拡声音到達時刻Ｔｏ２との和から、受音点Ｐｇ２において周辺スピーカＳｇ２からの拡声音が到達すべき時刻Ｔ２２（＝Ｔｏ２＋Δｔ２）を求める。拡声音の到達すべき時刻Ｔ２２が分かれば、周辺スピーカＳｇ２から受音点Ｐｇ２までの距離Ｒ２２を（１２）式へ代入することにより、ステップ６０８において周辺スピーカＳｇ２の発音時刻を決定できる。
【００５２】
ステップ６０９では、受音点Ｐｇ２において周辺スピーカＳｇ２からの拡声音到達時刻Ｔ２２と周辺スピーカＳｇ１からの拡声音到達時刻Ｔ１２とを比較し、周辺スピーカＳｇ１からの拡声音到達時刻Ｔ１２が先行する場合は、受音点Ｐｇ２における周辺スピーカＳｇ１からの到達拡声音の自乗音圧Ｉ１２と主スピーカＳｏからの到達拡声音の自乗音圧Ｉｏ２との和（Ｉ１２＋Ｉｏ２）を求め、その自乗音圧の和（Ｉ１２＋Ｉｏ２）を（１４）式へ代入することにより、周辺スピーカＳｇ１からの拡声音到達音圧レベルＬ１２と主スピーカＳｏからの拡声音到達音圧レベルＬｏｘとの総和（＝１０ｌｏｇ（Ｉ１２＋Ｉｏ２））を算出する。
【００５３】
本発明者は、各受音点Ｐｇｘにおいて下位順位の周辺スピーカＳｇｋ（ｋ＝１、２、．．．．．．、（ｘ−１））からの拡声音が周辺スピーカＳｇｘからの拡声音に対し先行して到達している場合は、主スピーカＳｏからの拡声音到達音圧レベルＬｏｘのみではなく、下位順位の周辺スピーカＳｇｋからの拡声音到音圧レベルＬｋｘ（＝１０ｌｏｇΣＩｋｘ）と主スピーカＳｏからの拡声音到達音圧レベルＬｏｘ（＝１０ｌｏｇＩｏｘ）との総和（＝１０ｌｏｇ（ΣＩｋｘ＋Ｉｏｘ））に対して先行音効果を与えるように当該受音点Ｐｇｘの上方の周辺スピーカＳｇｘからの拡声音到達音圧レベルＬｘｘを定めることにより、広い受音面Ｐｈにおいて、その上のあらゆる受音点で確実に先行音効果が得られることを見出した。本発明はかかる知見に基づくものである。
【００５４】
ステップ６１０では、例えば図８の関係式へ周辺スピーカＳｇ２の発音時刻の遅延時間Δｔ２を代入（ΔＬ２＝ｆ（Δｔ２））することにより音圧レベル差ΔＬ２を求める。求めたレベル差ΔＬ２を前記総和（＝１０ｌｏｇ（Ｉ１２＋Ｉｏ２））ヘ加えることにより、周辺スピーカＳｇ２からの到達音が受音点Ｐｇ２で示すべき音圧レベルＬ２２（＝１０ｌｏｇ（Ｉ１２＋Ｉｏ２）＋ΔＬ２）を求める。到達音圧レベルが求まれば、ステップ６１２において、上述した周辺スピーカＳｇ１の場合と同様に（１３）式への代入により周辺スピーカＳｇ２の入力パワーＷ２を決定できる。
【００５５】
ステップ６１２で周辺スピーカＳｇ２の発音時刻と入力パワーとを決定したのち、ステップ６１３から反復してステップ６０５へ戻り、周辺スピーカＳｇ３、Ｓｇ４についてステップ６０５〜６１２を繰り返す。図５のステップ６０５〜６１２を全ての周辺スピーカＳｇｘについて順位の昇順に繰り返すことにより、全ての周辺スピーカＳｇｘの発音時刻及び入力パワーが決定できる。コンピュータ３０により決定した各周辺スピーカＳｇｘの発音時刻及び入力パワーの指示信号を信号調整装置２０ヘ入力することにより、信号伝送装置１０からの音響信号の発音時刻と入力パワーとをスピーカＳｊ毎に調整し、調整後の音響信号をパワーアンプ１２経由で各スピーカＳｊへ入力する（図１参照）。
【００５６】
図９は、上述したように決定した各スピーカＳｏ、Ｓｇ１〜Ｓｇ４からの拡声音が受音点Ｓｇ４につくる音圧を示す。すなわち、この場合は話者Ｏからの肉声が最先の先行音として到達し、話者Ｏに最も近い主スピーカＳｏからの拡声音が続き、次に周辺スピーカＳｇ１、Ｓｇ２、Ｓｇ３の拡声音が順に到達する。最後に周辺スピーカＳｇ４からの拡声音が到達する。肉声と各スピーカからの拡声音との関係、スピーカからの拡声音の相互関係は先行音効果が働く条件を満たしているため、受音点Ｓｇ４の聴者２の音場知覚にとっては、合成された音圧の主観レベルが図１０の実線のようになると考えられる。すなわち図１０の実線は、複数の分散配置のスピーカからの拡声音が一つの音として聴者２に知覚されることを示す。
【００５７】
本発明は、実質的に同一の音響特性のスピーカを用いるのでスピーカの音色の違いはなく、またスピーカ間隔の適当な設定により聴者の両耳に対する伝送特性の差を許容範囲に収め、しかも各周辺スピーカの発音時刻と入力パワーを主スピーカからの拡声音が先行音となるように制御するので、受音面上のあらゆる位置で音源Ｏヘ向かう方向に音像が定位できる。また音源の位置の移動に応じて主スピーカの選択と周辺スピーカの順位付けを行なうことができるので、移動する音源に音像を定位させることができる。また不特定な位置に表れる質問者等に音像を定位させることも可能である。
【００５８】
こうして本発明の目的である「音像定位の方向を移動する音源の方向に合わせて変化させ得る話者移動に追従して音像定位する分散拡声方法及び分散拡声システム」の提供を達成できる。
【００５９】
好ましくは、各受音点Ｐｉの聴取音圧レベルを所定目標音圧レベルＬｈｃに抑制することにより、聴者２が受音面Ｐｈの如何なる受音点Ｐｉにいても、所定目標音圧レベルで聴取できるようにする。この場合は図５のステップ６１１で、例えば受音点Ｐｇ４における聴取音圧レベルを、受音点Ｐｇ４における周辺スピーカＳｇ４からの拡声音到達音圧レベルＬ４４（＝１０ｌｏｇＩ４４）と、先行して到達する下位順位の周辺スピーカＳｇｋ（ｋ＝１、２、３）及び主スピーカＳｏからの拡声音到達音圧レベルの総和（＝１０ｌｏｇ（ΣＩｋ４＋Ｉｏ４））との総計（＝１０ｌｏｇΣ（Ｉ４４＋ΣＩｋ４＋Ｉｏ４））として求める。求めた聴取音圧レベルが目標音圧レベルＬｈｃを超えるときは、周辺スピーカＰｇ４の到着音音圧レベルＬ４４を先行音効果が得られる範囲内において総計（＝１０ｌｏｇΣ（Ｉ４４＋Ｉｋ４＋Ｉｏ４））が目標音圧レベルＬｈｃと一致するまで抑制する。
【００６０】
更に好ましくは、目標音圧レベルＬｈｃを図１１の実線に示すように定め、各スピーカＳｊのつくる受音面Ｐｈ上の合成音場の音圧レベルが、音源Ｏから離れるに従って減衰するように制御する。この減衰制御は、音源Ｏから離れると音が小さくなるという日常の経験と合致した音場の形成を可能とし、聴者にとって音場が自然な感じで知覚されるように作用する。このため、講演会などであれば聴者は話者から話しかけられているという感じをもちながら聴くことができ、話者は聴衆の反応をよく感じ取りながら話題を展開する等の心理的なフィードバックが可能となリ、参加者にとって互いの意思の疎通が図れるという効果が期待できる。
【００６１】
図１１に示す目標音圧レベルＬｈｃは、音源Ｏからの距離Ｒｈに応じてレベルが低下するが、距離Ｒｈに対する低下の傾きが自由音場における音の減衰の傾きよりも小さいものである。この目標音圧レベルＬｈｃによれば、聴者に自然な知覚を与えつつ、十分大きな音圧で拡声することができる。図中の４０ｄＢ以下の黒く塗りつぶした部分は、音が小さすぎて聴者の聞き取りに支障が生じる範囲を示す。自由音場では、話者がやや大きい声で発声した場合でも１００ｍ程度離れた聴者には聞き取り難い場合があるが、図１１の目標音圧レベルＬｈｃを用いた拡声方法によれば、聴者に自然な音場感覚を与えつつ、１００ｍ離れた地点でも十分な音圧で話者の発声を聞くことができる。
【００６２】
本発明者は、図１１に示すように目標音圧レベルＬｈｃとして、主スピーカＳｏ下方の受音点Ｐｏにおける主スピーカＳｏからの拡声音到達音圧レベルＬｏｏを最大レベルとし、主スピーカＳｏからの距離Ｒｈの対数（ｌｏｇＲｈ）に比例してレベルが低減するような傾斜型目標音圧レベルＬｈｃ＝Ｌｏｏ−３．０×ｌｏｇＲｈを用いることにより、聴者が音と距離との不自然なずれを感じることなく、話者から離れた位置でも話者が近くにいるかのように明瞭でスッキリした自然な音を聞くことができる音場が形成できることを実験的に確認した。ただし、傾斜型目標音圧レベルＬｈｃの減衰の比例定数は図示例に限定されない。
【００６３】
【実施例】
図１２は信号調整装置２０とアンプ２４を一体にした複合アンプ２０ｂを用いた本発明の実施例を示し、図１３は同実施例で使用する複合アンプ２０ｂのブロック図を示す。図示例では、複合アンプ２０ｂの相互間はタンデム接続されており、１本のデジタル音響信号ライン２７と制御信号ライン２８とにより複合アンプ２０ｂの各々へ音響信号と制御信号とを供給している。複合アンプ２０ｂは２つの音響遅延部２１Ｌ、２１Ｒと２つの音量制御部２２Ｌ、２２Ｒを有し、２チャンネルのデジタル音響信号を同時に処理することができる。
【００６４】
コンピュータ３０（図１４参照）により算出された主スピーカＳｏ及び各周辺スピーカＳｇｘの発音時刻（遅延時間）と入力パワー（音量）は、インターフェース１４及び制御ライン２８経由で複合アンプ２０ｂのＣＰＵ２５へ送られる。ＣＰＵ２５は音響遅延部２１Ｌ、２１Ｒ、音量制御部２２Ｌ、２２Ｒとに接続され、パターンメモリ２６に記憶された処理パターンに従って２チャンネルのデジタル音響信号の発音時刻及び入力パワーを調整する。調整後のデジタル音響信号は、ミキサ２３でミキシングされ、アンプ２４、出力ライン２９経由でスピーカへ送られる。同図の複合アンプ２０ｂを用いれば、２つの音源の音を同時に処理することができ、受音面Ｐｈ上の２箇所の音源に向けた音像を同時に定位させることが可能となる。
【００６５】
図１４は、複合アンプ２０ｂを用いた本発明の更に他の実施例を示す。床面から３．５ｍの高さに５ｍ×３．５ｍの間隔で格子状に１４台のスピーカＳｊを分散配置し、各スピーカＳｊにそれぞれ複合アンプ２０を取り付けた。同図に示すように、スピーカＳ１の下方の第１音源とスピーカＳ４の下方の第２音源の音をそれぞれマイクロホン３及び４で収音し、ミキサー１１経由でインターフェース１４へ入力した。他方、コンピュータ３０では２つの音源への音像定位に必要な制御信号を図６の流れ図に従ってそれぞれ算出し、インターフェース１４へ入力した。ミキサ１１からのデジタル音響信号とコンピュータ３０からの制御信号とをインタフェース１４経由で各スピーカＳｊの複合アンプ２０へ伝送した。各複合アンプ２０は図１４に示すようにタンデム状に接続した。複合アンプ２０ｂにおいて右チャンネルヘ送られたマイクロホン３からのデジタル音響信号と、左チャンネルへ送られたマイクロホン４からのデジタル音響信号とを同時に処理し、ミキサー２３及びメインアンプ２４経由で各スピーカＳｊへ出力した。
【００６６】
図１４の実施例で用いた制御信号の一例を表２に示す。本発明者は、表２の制御信号により、受音面Ｐｈ上で第１音源へ向かう音像と第２音源ヘ向かう音像とを同時に定位できることを確認した。図１４及び表２の実施例から分かるように、例えば舞台上の講演者と会場内の質間者などのように音場内に２つの音源Ｏが存在する場合でも、本発明の拡声方法によれば、講演者を第一音源として音像定位し且つ同時に質間者を第二音源として音像定位を行うことが可能である。この同時に２つの音像定位を行なうことにより、講演者と質問者がお互いの声をよく聞くことができるので話しやすい、さらには会場内のどこにいても両者のやりとりが自然な感じで聞こえる、という音場を創り出すことができる。
【００６７】
【表２】

【００６８】
【発明の効果】
以上詳細に説明したように、本発明に係る話者移動に追従して音像定位する分散拡声方法及び分散拡声システムによれば、受音面上方に実質上同一音響特性の複数のスピーカＳｊをカバーエリアを一部重畳させて下向きに分散配置し、各スピーカＳｊの分散配置された音場内にて移動する音源Ｏの位置を検出し、各スピーカＳｊのうち検出した音源Ｏの位置に最も近いスピーカを主スピーカＳｏとし且つ残余のスピーカを主スピーカＳｏからの距離の昇順に順位付けされた周辺スピーカＳｇｘとし、主スピーカからの拡声音が先行音効果を与えるように各周辺スピーカの発音時刻と入力レベルを制御し、受音面上の音像の定位方向を音源Ｏヘ向かう方向とするので、音源の位置の変化に対応することができる。すなわち、音源が移動する場合であっても主スピーカを選択し直すことにより、移動に応じた音源へ向かう方向に音像を定位することができる。したがって、話者の位置が変化する場合、例えば話者が広い会場を移動する場合或いは舞台位置が変化する場合においても、話者の位置に合わせて主スピーカＳｏを実質的に配置許容範囲内に移動させたのと同等の状態が得られることになる。これにより、音源位置の移動や質問者の不特定な位置での出現に応じて音像定位の方向を追従させることができ、大きな講演会場等において、移動する話者の位置や質問者の位置が分からない等の問題を解決することができる。
【図面の簡単な説明】
【図１】本発明に係る実施の形態の説明図である。
【図２】インテリジェントパワーアンプとタグ読み取り装置との接続例の説明図である。
【図３】スピーカの音響特性の一例の説明図である。
【図４】先行音効果を得るための先行音スピーカの配置許容範囲の説明図である。
【図５】各スピーカの発音時刻と入力パワーの算出方法の流れ図の一例である。
【図６】本発明の音像定位の原理を示す説明図である。
【図７】先行音効果の説明図である。
【図８】先行音効果が得られる先行音と補強音との関係の説明図である。
【図９】図６の受音点における各スピーカからの到達音の説明図である。
【図１０】図６の受音点における音強度の説明図である。
【図１１】受音点の目標音圧レベルの説明図である。
【図１２】２チャンネル複合アンプを用いた本発明の実施例の説明図である。
【図１３】２チャンネル複合アンプの説明図である。
【図１４】本発明の他の実施例の説明図である。
【図１５】従来の集中拡声方式、分散拡声方式、及び半分散拡声方式の説明図である。
【図１６】従来の集中拡声方式、分散拡声方式、及び半分散拡声方式における受音点での到達音声レベルの説明図である。
【符号の説明】
３…マイクロホン、１０…信号伝送装置、２０…信号調整装置、３１…スピーカ選択手段、３２…スピーカ順位付け手段、３３…主スピーカ音指示手段、３４…周辺スピーカ音指示手段、３７…マイクロホン位置検出装置、１０１…赤外線発信タグ、１０３…タグ読み取り装置、１１０…音源位置同定手段、Ｏ…音源、Ｐｈ…受音面、Ｐｉ…受音点、Ｓｊ…スピーカ、Ｓｏ…主スピーカ、Ｓｇｘ…周辺スピーカ、Ｐｇｘ…下方の受音点[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a distributed loudspeaker method and a distributed loudspeaker system for localizing a sound image following speaker movement, and more particularly to a technique for vocalizing voice and performance in a space used for gatherings such as lectures, ceremonies, and conferences.
[0002]
[Prior art]
Hearing the voice of the speaker from the direction of the speaker (sound source) (sound image localization) has the effect of increasing the sense of presence and maintaining the concentration of the listener. Conventional loudspeaker systems include, for example, a centralized loudspeaker system and a distributed arrangement loudspeaker system. FIG. 15A shows a centralized loudspeaker system as an example of a conventional loudspeaker system, and FIG. 15B shows a distributed arrangement loudspeaker system as another example. In the centralized loudspeaker system, a microphone is installed in the vicinity of a sound source O such as a speaker or a performer, and a sound pressure change caused by vocalization or the like is amplified by a microphone amplifier and a power amplifier and the like, and is amplified by one or more speakers S. The sound from the sound source O is transmitted to the listener 2. The speakers S are usually arranged on both sides of the stage. In the distributed arrangement loudspeaker system, the sound of the sound source O is simultaneously loudspeaked from a plurality of speakers S0 to S4 arranged at fixed intervals on a ceiling or the like.
[0003]
FIGS. 16A and 16B show the time when the loud sound from each speaker reaches a sound receiving point (hereinafter, referred to as a loud sound arrival time) and the sound receiving point in the centralized loudspeaker system and the distributed loudspeaker system. An example of the relationship with the sound pressure level of the loudspeaker sound (hereinafter, referred to as loudspeaker arrival sound pressure level) is shown.
[0004]
However, the conventional centralized loudspeaker system and the distributed arrangement loudspeaker system have the following problems.
(1) The viewing direction is different from the audible direction.
In the centralized loudspeaker system in which the speakers are on both sides of the stage, when the seat position is on the right side of the symmetry axis at the center of the stage, the voice of speaker O is heard from the right speaker and the speaker O comes to the center of the visual field. On the left side, the voice of speaker O can be heard from the left speaker. Further, in the distributed placement and loudspeaker system, the sound of the speaker O on the stage can be heard from the overhead speaker. Both methods have a problem that the listener 2 has a different recognition direction of the speaker O for vision and hearing.
[0005]
(2) The sound is out of focus.
In the distributed arrangement loudspeaker system, as shown in FIG. 16B, the sound reaching the sound receiving point from the plurality of speakers S0 to S4 has a difference in direction and time, and interference of sound waves from the speakers S0 to S4 occurs. As a result, many peaks and valleys occur in the frequency characteristics at the sound receiving point. As a result, the sound of the sound source is blurred and unclear at the sound receiving point.
[0006]
(3) The volume of the sound becomes unnatural.
In the concentrated diffusion method, by disposing the speaker S near the center of the stage, it is possible to eliminate the mismatch between the visual and auditory directions. However, the intensity of the sound is significantly different between a place close to the stage and a place far from the stage, and there is a problem that the sound is noisy near and inaudible at a distance. On the other hand, if the distributed speakers are all loudspeaked at the same volume, a uniform sound field can be obtained, but there is a problem that the listener feels that the speaker's appearance and sound volume are unnatural. If the sound is heard at the same volume as the close position even at a position away from the speaker, the sound is unnatural because it is significantly different from the everyday experience, that is, the experience that the voice decreases when the speaker is away from the speaker. It is said that this unnaturalness impairs the feeling (interactivity) being spoken by the speaker, and produces results such as being unable to concentrate on the talk and not communicating what they want to say.
[0007]
As a loudspeaker method for solving the above-mentioned problem, as shown in FIG. 15C, a main sound source speaker (main speaker) S0 is arranged near a sound source O, and distributed speakers S1 to S1 are placed on a ceiling remote from the sound source. There is a semi-variable loudspeaker system in which S3 is arranged. In this method, in order to obtain a feeling that sound comes from the speaker direction even at a sound receiving point close to the distributed speaker (sound image localization in a direction coincident with the speaker direction), the sounding time of the distributed speaker is compared with the sounding time of the main speaker. Is controlled to delay. FIG. 16C shows an example of the relationship between the arrival time and the arrival sound pressure level at the sound receiving point of the loudspeakers from the speakers S0 to S3 according to the half-dispersion loudspeaker system.
[0008]
The direction of sound image localization is based on human auditory perception called the precedence effect. As shown in FIG. 7, for example, as shown in FIG. 7, when a sound (preceding sound) from the direction of the speaker A is heard at a sound receiving point, a sound with a certain range of time delay and a certain range of sound is heard from the speaker B in another direction. This is an psychoacoustic effect in which another sound (reinforcement sound) that arrives with a pressure level difference is localized in the direction of the speaker A (the direction of the preceding sound). FIG. 8 shows the arrival time difference between the preceding sound and the reinforcement sound at which the preceding sound effect is obtained at the sound receiving point (delay time, Tb-Ta = reception sound arrival point arrival time of the reinforcement sound−preceding sound arrival point arrival) The relationship between (time) and the reached sound pressure level difference (Lb-La = sound pressure level at the sound receiving point of the reinforcement sound-sound pressure level at the sound receiving point of the preceding sound) is shown. The range indicated by oblique lines in the drawing is the range where the preceding sound effect can be obtained, that is, the sound image localization region. The precedence effect may be called the Haas effect or the first wavefront law.
[0009]
In the half-dispersion loudspeaker system, as shown in FIG. 16C, control is performed so that the sound from the main speaker S0 reaches the listener 2 at the sound receiving point first. The reinforcement sound from the distributed speakers S1 to S3 (hereinafter, sometimes referred to as a succeeding sound) has a time delay of about 10 ms from the preceding sound from the main speaker S0 to the listener 2 at the sound receiving point, and a certain range shown in FIG. Is controlled so as to arrive with the difference in sound pressure level. By controlling the arrival time and arrival sound pressure level of the loud sound from each of the speakers S0, S1 to S3 in this manner, it is possible to localize the sound image in the direction of the main speaker at the sound receiving point.
[0010]
[Problems to be solved by the invention]
However, the half-dispersion loudspeaker method has a problem that it cannot cope with a change in the position of a sound source. In order to obtain the sound image localization, it is necessary that the difference between the direction of the speaker (sound source) viewed from the listener and the direction of the preceding sound speaker is within a certain allowable range. Referring to FIG. 4, the speaker (sound source) and the listener (sound receiving point) are on opposite sides (positive side and negative side) of the origin on the Y axis on the XY plane (horizontal plane), and the preceding sound Assuming that the speaker is on the XZ plane, a range of ± 10 degrees in the X axis direction and ± 45 degrees in the Z axis (vertical) direction with respect to the speaker direction (Y axis direction) when viewed from the listener (see FIG. The range indicated by oblique lines) is the allowable range of the preceding sound speaker. When the position of the speaker changes in the semi-dispersed loudspeaker system, for example, when the speaker moves in a large hall or when the stage position changes, the main speaker S0 is moved to the position shown in FIG. Although it is necessary to move within the range, it is actually difficult to move the main speaker S0. Further, even when a questioner or the like appears at an unspecified position in the audience seat, it is impossible to localize the sound image to the questioner. For this reason, there have been problems such as that the position of the questioner cannot be determined in a large lecture hall or the like. It is desired to develop a loudspeaker method that can follow the direction of sound image localization in accordance with the movement of the sound source position or the appearance of an interrogator at an unspecified position.
SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and provides a distributed loudspeaker method and a distributed loudspeaker system for localizing a sound image following speaker movement that can change the direction of sound image localization in accordance with the direction of a moving sound source. Is to do.
[0011]
[Means for Solving the Problems]
Referring to the embodiment of FIG. 1, the distributed loudspeaker method for localizing a sound image following speaker movement according to the present invention is substantially the same sound used for loudspeaking the sound of a sound source O at a predetermined position above a sound receiving surface Ph. A plurality of loudspeakers Sj (1 ≦ j ≦ n) having characteristics are dispersed and arranged downward with a part of the cover area of the adjacent loudspeaker partially overlapped, and a portion on the sound receiving surface Ph vertically below each loudspeaker Sj is received. A sound point Pi (1 ≦ i ≦ n) is set, and the position of the sound source O moving in the sound field in which the speakers Sj are dispersed is detected. Peripheral speakers Sgx (x = 1, 2,..., (N−1)) in which a nearby speaker is selected as the main speaker So and the remaining speakers are ranked in ascending order of distance from the main speaker So. From the main speaker So The loudspeaker of the sound of the source O is produced at the required input power at time zero with no time delay, and each of the peripheral speakers Sgx is time-delayed in the ascending order of the order and produced at the required input power, thereby producing each of the receiving speakers. The localization direction of the sound image at the sound point Pi is the direction toward the sound source O.
[0012]
Further, referring to the embodiment of FIG. 1, the distributed loudspeaker system according to the present invention for localizing a sound image following a movement of a speaker partially overlaps a cover area of an adjacent speaker at a predetermined position above a sound receiving surface Ph. Loudspeakers Sj (1 ≦ j ≦ n) having substantially the same acoustic characteristics, which are distributed downward, and the position of a sound source O moving in a sound field in which each of the speakers Sj is distributed is detected and a detection signal is transmitted. A microphone position detecting device, a signal transmitting device for transmitting an acoustic signal from a sound source O to each of the speakers Sj, a signal transmitting device provided between the signal transmitting device and each of the speakers Sj, and for each of the speakers Sj according to an instruction input. A signal adjusting device that adjusts a sounding time and an input power of the acoustic signal and outputs the signal to each of the speakers Sj; Speaker selection means for selecting the speaker closest to the sound source O as the main speaker So, and speakers Sj other than the main speaker So as peripheral speakers Sgx (x = 1, 2,..., (N-1)) Speaker ranking means for ranking in ascending order of distance from the main speaker So, main speaker sound instructing means for indicating the sounding time and input power of the main speaker So, and sounding time and input power of the peripheral speaker Sgx. Peripheral speaker sound instructing means for inputting sounding time and input power instructions from the main speaker sound instructing means and the peripheral speaker sound instructing means to the signal adjusting device to localize the sound image at each sound receiving point Pi. As a direction toward the sound source O.
[0013]
Preferably, in the distributed loudspeaker system for localizing a sound image following a movement of a speaker according to claim 3 or 4, wherein the microphone position detecting device is provided in the microphone and transmits an tag identification number, and an infrared transmitting tag, and each of the speakers Sj. A tag reading device that is provided at the same time and has a reader identification number, reads the tag identification number from the infrared transmitting tag that has entered the cover area of each speaker Sj, and sends out the tag identification number and the reader identification number; Sound source position identification means for detecting an arbitrary position of the microphone moving in the field based on the tag identification number and the reader identification number.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 shows an embodiment in which 25 speakers Sj (1 ≦ j ≦ 25) are dispersed and arranged downward above the sound receiving surface Ph. The sound receiving surface Ph is a virtual surface assumed at, for example, the height position h of the ear of the listener 2 on the floor surface 6. Although FIG. 3 shows a horizontal sound receiving surface Ph parallel to the floor surface 6, the sound receiving surface Ph targeted by the present invention is not limited to a horizontal plane. It is desirable from the viewpoint of signal adjustment control described later that the speakers Sj are disposed above the sound receiving surface Ph at a fixed height and downward so that their mutual intervals are constant.
[0015]
However, the distributed arrangement of the loudspeakers Sj of the present invention is not necessarily limited to the distributed arrangement at a fixed height and at an interval. For example, when there is a step on the ceiling, the height from the sound receiving surface Ph to each speaker Sj is not constant, but by changing the input power of each speaker Sj according to the difference in height, the sound receiving surface Ph is changed. The difference in sound pressure of the loudspeaker above can be compensated. Further, as described later, if the overlap of the cover areas of the adjacent speakers Sj can be obtained, for example, about 20% or more, the intervals between the speakers Sj do not need to be constant.
[0016]
Each speaker Sj is connected to the computer 30 via an intelligent power amplifier IPA described later. A microphone position detecting device 37 is connected to the computer 30. The microphone position detecting device 37 includes an infrared transmitting tag 101 serving as a position signal sending unit provided in the microphone 3, a tag reading device 103 serving as a position signal reading unit provided in each speaker Sj, and a program stored in the computer 30, for example. And a certain sound source position identification means 110.
[0017]
The infrared transmission tag 101 has a power source such as a battery, a transmission circuit, and a memory, and transmits an assigned unique tag identification number. The tag reading device 103 can receive the signal of the tag identification number transmitted from the infrared transmission tag 101 together with the detection value of the reception intensity. Therefore, for example, even when the microphone 3 enters the overlapping area of the cover areas of two adjacent speakers Sj, Sj, it is possible to detect which speaker Sj, Sj is closer from the received signal strength. .
[0018]
Each tag reading device 103 has a unique reader identification number. The tag reading device 103 reads the tag identification number from the infrared transmission tag 101 that has entered the cover area of the speaker Sj, and sends the tag identification number and the reader identification number assigned to itself to the computer 30.
[0019]
Each speaker Sj has an intelligent power amplifier IPA. The intelligent power amplifier IPA includes communication means such as a CPU, a memory, and an RS485, and has a function of receiving a command signal from the outside, calling a memory, setting parameters, performing volume adjustment, and the like. As shown in FIG. 2, the tag reading device 103 is connected to the signal line 109 via the interface 105, and is connected to the computer 30 via the interface 111. The intelligent power amplifier IPA is also connected to the signal line 109 via the interface 107 and to the computer 30 via the interface 111.
[0020]
Based on the tag identification number from the infrared transmission tag 101 read by the tag reading device 103 and the reader identification number of the tag reading device 103, the microphone 3 It is possible to detect whether it exists in the cover area. As shown in FIG. 1, when the microphone 3 is at the position M1, the sound source position identification means 110 detects that the microphone 3 is located in the cover area of the speaker S1, and when the microphone 3 is at the position M2, the cover area of the speaker S3. Is located by the sound source position identification means 110. The computer 30 sends a setting signal of each parameter to the intelligent power amplifier IPA of each speaker Sj based on the result of the sound source position identification means 110.
[0021]
In FIG. 1, a plurality of speakers Sj having substantially the same acoustic characteristics are used, and are arranged at a height H where a part of a cover area of an adjacent speaker, for example, about 20% overlaps on the sound receiving surface Ph. Rather than using a main speaker different from a distributed speaker as in the half-dispersion loudspeaker system, the timbre from each speaker Sj is obtained by using speakers Sj having substantially the same acoustic characteristics such as sound pressure frequency characteristics and directivity coefficients. To prevent the preceding sound effect from being disturbed due to the difference in timbre on the sound receiving surface Ph. Further, for example, by partially overlapping the cover area of one octave band centered at 1 kHz, preferably a band of 500 Hz to 2 kHz, it is possible to avoid the occurrence of a region where the sound of the band does not reach on the sound receiving surface Ph, It is possible to obtain a preceding sound effect at any position on the sound receiving surface Ph.
[0022]
In general, the characteristic A0 (f) of the speaker Sj is a point 1 m on the speaker front axis (horizontal direction 0 °, vertical direction 0 °) when a 1 (watt) signal is input at a frequency f (Hz). It is defined as the squared sound pressure P2 (the unit is the square of Pa (Pascal), that is, Pa2). The sound pressure level L is defined as a squared sound pressure P2 based on a sound pressure P0 (Pa) of 20 μPa as shown in the following equation (1). In the following description, the squared sound pressure P2 (P2 / P02) based on the sound pressure P0 of 20 μPa is referred to as the squared sound pressure I.
[0023]
The sound pressure frequency characteristic of the speaker is defined as a frequency characteristic of a sound pressure level L at a point 1 m of the front axis of the speaker when a 1 (watt) signal is input. If the sound pressure in the horizontal directivity direction φ and the vertical directivity direction θ is represented by A (f, θ, φ), and A0 (f) is represented by A (f, 0, 0), the directivity coefficient D (f, θ) , Φ) are defined as in the following equation (2). The squared sound pressure I and sound pressure level L of the loudspeaker reaching the sound receiving point at a distance R (m) away from the speaker in the horizontal directivity direction φ and the vertical directivity direction θ are expressed by the following equations (3) and (3). As shown in Equation 4), the distance R (m) from the speaker S to the sound receiving point, the directivity coefficient D (f, θ, φ) of the speaker, the sound pressure frequency characteristic A (f) of the speaker, and the input It is expressed as a function of the power W (watt).
[0024]
(Equation 1)

[0025]
In FIG. 1, only speakers Sj having the same acoustic characteristics such as sound pressure frequency characteristics and directivity coefficients are used. The graph of FIG. 3B shows a directional frequency characteristic of an opening angle (directivity angle) at which the sound pressure level L attenuates by −6 dB with respect to the front axis. For example, when the speaker having the directional frequency characteristics shown in FIG. 3B is selected as the distributed speaker Sj of the present invention, since the directional angle of 1 kHz is about 85 to 90 degrees, the directional angle and the superposition ratio of the cover area are determined. The arrangement height of each speaker Sj from the sound receiving surface Ph can be designed in consideration of the above. The speaker arrangement shown in FIG. 1 is an example in which speakers having the directional frequency characteristics shown in FIG. 3 are distributed.
[0026]
However, the speakers Sj used in the present invention are not necessarily limited to the same type. For example, when there is little difference in sound quality between one octave band centered on 1 kHz, preferably a band of 500 Hz to 2 kHz, and the sound characteristics are substantially the same, different types of speakers can be used.
[0027]
The mutual distance between the distributed speakers Sj is determined so that the transmission characteristics of the two speakers Sj in the sound source O direction with respect to the sound receiving point are similar. As will be described later, in the present invention, the sound image is localized with the speaker Sj closest to the sound source O as the preceding sound speaker, and the surrounding speakers Sj in the direction from the sound source O to the sound receiving point (listener 2) as the subsequent sound speakers. By appropriately designing the distance between the distributed speakers Sj in consideration of the relationship with the arrangement height, at any sound receiving point on the sound receiving surface Ph, the sound of the preceding sound speaker and the following sound speaker in the sound source direction with respect to both ears. The difference in transmission characteristics can be set within an allowable range in which sound image localization is possible.
[0028]
Further, in the above-described arrangement of the speakers Sj, the deviation between the direction of the sound source O and the direction of the preceding sound speaker as viewed from the sound receiving point falls within the allowable range shown in FIG. For example, in FIG. 1, when the speaker O is below any of the other speakers Sj when viewed from the listener 2 below the speaker S5, the speaker Sj above the speaker with respect to the direction of the speaker O in FIG. Since the speaker is located within the allowable range, a sound image localization can be obtained using the speaker above the speaker as a preceding sound speaker. That is, in the present invention, the plurality of speakers Sj are dispersedly arranged at predetermined positions which satisfy the height at which the cover area is superimposed, the mutual interval at which similar transmission characteristics are obtained, and the allowable range of the preceding sound speakers.
[0029]
An acoustic signal from a sound source O is input to the dispersed speakers Sj by the signal transmission device 10. The signal transmission device 10 of FIG. 1 includes a microphone 3, a mixer 11, and a signal transmission path. The microphone 3 is disposed near a speaker, a performer, or another sound source O in a sound field to directly collect sound. The collected direct sound is amplified by a microphone amplifier (not shown), and then the volume of the entire sound field is controlled by the mixer 11 and sent to a signal adjusting device 20 described later. The microphone volume control in the mixer 11 can be manually controlled by an operator in consideration of the volume of the sound source O. However, the scope of application of the present invention is not limited to the loudspeaking of the sound of the sound source O in the sound field, but can be applied to, for example, a case where a sound image is localized by loudspeaking from a previously recorded sound source.
[0030]
In the present invention, a signal adjustment device (effector) 20 for adjusting the acoustic delay and the sound pressure level of each speaker Sj is inserted between the signal transmission device 10 and each speaker Sj. As shown in FIG. 13, an example of the signal adjustment device 20 is an audio delay that adjusts the sounding time and input power of an audio signal for each speaker Sj in response to input of an instruction signal from a computer 30 (see FIG. 14). It has a circuit 21 and a sound pressure level control circuit (volume control circuit) 22. Each signal conditioner 20 is connected to a computer 30, and the computer 30 controls the acoustic delay and the input power of each signal conditioner 20 collectively.
[0031]
FIG. 5 shows an example of a flow chart of the control of the signal adjustment device 20 by the computer 30 in the case where the sound image localization is obtained by loudspeaking the sound of the sound source O in the sound field. Hereinafter, the loudspeaker method of the present invention will be described with reference to the flowchart of FIG. First, in step 601, the position of the sound source O in the sound field is detected. In FIG. 1, the position of the sound source O is detected based on the detection signal of the microphone position detecting device 37.
[0032]
That is, a tag identification number is transmitted from the infrared transmission tag 101 attached to the microphone 3. This tag identification number is detected by the tag reading device 103 in the cover area where the microphone 3 is located. The tag reading device 103 sends the tag identification number and the reader identification number assigned to itself to the computer 30. The computer 30 detects the position of the microphone 3 from these identification numbers by the sound source position identification means 110. That is, the position of the microphone 3 is automatically detected by the microphone position detecting device 37.
[0033]
Next, in step 602, the speaker S1 closest to the sound source O is selected by the speaker selection means 31 as the main speaker So. An example of the speaker selection means 31 is a computer built-in program that calculates the distance between the sound source O and each speaker Sj and selects the speaker S1 having the shortest distance.
[0034]
In step 603, the distributed speaker Sj other than the main speaker So is ranked in ascending order of the distance from the main speaker So by the speaker ranking means 32 which is a program built in the computer 30, for example. In this specification, the ranked speakers Sj are called peripheral speakers Sgx (x = 1, 2,..., (N-1)). For example, in FIG. 1, the speakers S2 in the distributed arrangement are the peripheral speakers Sg1 closest to the sound source O, the speakers S3 are the peripheral speakers Sg4 closest to the fourth, the speakers S4 are the peripheral speakers Sg8 closest to the eighth, and the speakers S5 are 15 It is ranked as the nearest neighboring speaker Sg15.
[0035]
As shown in steps 601 to 603, in the present invention, the position of the main speaker So is not fixed, the distributed speaker closest to the sound source O is set as the main speaker So, and the peripheral speakers Sgx are ranked according to the distance from the main speaker So. Therefore, even when the sound source O moves, the position of the main speaker So and the order of the surrounding speakers Sgx are changed according to the movement, and the sound image localization on the sound receiving surface Ph is changed. That is, the sound image can be localized following the moving sound source O.
[0036]
In step 604, the sounding time To and the input power Wo of the main speaker So are instructed by the main speaker sound instructing means 33 which is a program in the computer 30, for example. For example, an acoustic signal from the signal transmission device 10 is generated by the main speaker So without time delay, and the sound generation time To is set to time zero (ms). However, the sound image localization of the present invention can be obtained even if a time delay is provided to the sounding time To of the main speaker So. Also, the input power Wo of the main speaker So is such that the sound pressure level of the loudspeaker reaching the microphone 3 from the main speaker So is higher than the sound pressure level of the real voice from the speaker O (the sound pressure level at a point 1 m away). The level is determined to be a level required for preventing howling, for example, a level lower by 2 dB. However, when no howling occurs, the input power Wo of the main speaker So can be set to a required level. When determining the input power Wo of the main speaker So, the volume of the mixer 11 corresponding to the microphone 3 can be appropriately adjusted, and the amount of attenuation of the attenuator of the amplifier 12 (hereinafter, may be referred to as the attenuator position). Can be set to 0 (dB).
[0037]
The method of determining the sounding time and input power of each of the peripheral speakers Sgx by the peripheral speaker sound instructing means 34 after step 605 will be described by arranging five speakers S1 to S5 between the speaker O and the listener 2 for the sake of simplicity. A description will be given with reference to FIG. FIG. 6 illustrates a case where the speaker S1 closest to the speaker O is set as the main speaker So, and the speakers S2 to S5 are ranked as the peripheral speakers Sg1 to Sg4, and the listener 2 is located below the peripheral speaker Sg4. In step 605, first, sound receiving points P1 to P5 are determined on a sound receiving surface Ph vertically below the respective speakers S1 to S5. In FIG. 6, a sound receiving point P1 is a sound receiving point Po vertically below the main speaker So, and sound receiving points P2 to P5 are sound receiving points Pgx (Pg1 to Pg4) vertically below the peripheral speakers Sgx (Sg1 to Sg4). is there.
[0038]
The time required for the loud sound from the main speaker So to reach the sound receiving point Pgx below the peripheral speaker Sgx is represented by the following equation (10), where Rox is the distance between the main speaker So and the sound receiving point Pgx. Can be calculated. In the equation (10), the symbol C represents the speed of sound (m / s). Normally, a sound speed at normal temperature and normal humidity (23 ° C., relative humidity 60%) is used as the sound speed C. Also, the sound pressure level Lox of the loud sound from the main speaker So that has reached the sound receiving point Pgx can be expressed as the following equation (11) with reference to the equation (4). In the equation (11), Wo represents the input power of the main speaker So, and Do represents the directivity coefficient of the main speaker So.
[0039]
Also, the time required for the loud sound to reach from the surrounding speaker Sgx to the sound receiving point Pgx below the surrounding speaker Sgx and the reached sound pressure level Lxx are the same if the distance between the surrounding speaker Sgx and the sound receiving point Pgx is Rxx. Expressions (12) and (13) can be expressed in the same manner as in the case of the speaker So. In equation (13), Wx represents the input power of the peripheral speaker Sgx, and Dx represents the directivity coefficient of the peripheral speaker Sgx. In the illustrated example, since the main speaker So and the peripheral speaker Sgx have the same acoustic characteristics, A (f) in the equations (11) and (13) is the same.
[0040]
Loudened sounds from the main speaker and a plurality of peripheral speakers can reach each sound receiving point Pgx. A combined sound pressure level (hereinafter, listening sound pressure) at a sound receiving point Pgx at which loudspeakers arrive from a plurality of speakers Sj (j = 1, 2, 3,...) Having the same characteristic A (f). Is referred to as a level) and can be expressed by the following equation (14) based on the sum of the squared sound pressures I of the sound pressures reached from the respective speakers Sj.
[0041]
(Equation 2)

[0042]
In steps 605 to 612, the sounding time and input power of each of the peripheral speakers Sgx are determined according to the ascending order of the order of each of the peripheral speakers Sgx. First, for the peripheral speaker Sg1 in FIG. 6, in step 605, the distance Ro1 from the main speaker So to the sound receiving point Pg1 below the peripheral speaker Sg1 and the distance R11 from the peripheral speaker Sg1 to the sound receiving point Pg1 are calculated. In the case of the peripheral speaker Sg1, there is no lower-order peripheral speaker Sgx, that is, the peripheral speaker Sgx closer to the main speaker So than the peripheral speaker Sg1. In step 606, the main speaker So at the sound receiving point Pg1 is obtained from the substitution result of the distance Ro1 from the main speaker So to the sound receiving point Pg1 to the expression (10) and the sounding time To = 0 (ms) of the main speaker So. Is calculated. From the result of substituting the distance Ro1 from the main speaker So to the sound receiving point Pg1 and the input power Wo and the directivity coefficient Do of the main speaker So into the equation (11), the loud sound from the main speaker So at the sound receiving point Pg1. The attained sound pressure level Lo1 is calculated. However, the sound pressure frequency characteristic A (f) of the main speaker So is predetermined.
[0043]
In step 607, a delay time Δt1 for providing a preceding sound effect with respect to the loud sound arrival time To1 from the main speaker So is obtained. The delay time Δt1 is the shortest delay time Δtmin, for example, 3 ms for the peripheral speaker Sg1 closest to the main speaker So, the longest delay time Δtmax, for example, 10 ms for the peripheral speaker Sg4 farthest from the main speaker So. In the peripheral speakers Sg2 and Sg3, the time can be determined as a time proportionally distributed to the distances Ro2 and Ro3 from the main speaker So to the sound receiving points Pg2 and Pg3 below the peripheral speakers Sg2 and Sg3.
[0044]
However, the method for determining the delay time Δt1 is not limited to the method based on the proportional distribution according to the distance. For example, it can be determined as the time between the shortest delay time and the longest delay time proportional to the logarithm (logRox) of the distance Ro2, Ro3 from the main speaker So to the sound receiving points Pg2, Pg3 below the peripheral speakers Sg2, Sg3. . Table 1 below shows the delay time Δtx of each of the peripheral speakers Sg1 to Sg4 determined from this relationship. However, in Table 1, the delay time Δt1 of the peripheral speaker Sg1, which is the shortest delay time Δtmin, is 6 ms, and the delay time Δt4 of the peripheral speaker Sg4, which is the longest delay time Δtmax, is 30 ms.
[0045]
In step 607, the arrival time T11 (= To1 + Δt1) at which the loud sound from the peripheral speaker Sg1 should reach the sound receiving point Pg1 is calculated as the sum of the obtained delay time Δt1 and the loud sound arrival time To1 from the main speaker So. I do. When the time T11 at which the loudspeaker sound should arrive is known, the distance R11 from the peripheral speaker Sg1 to the sound receiving point Pg1 is substituted into the equation (12), so that at step 608 the peripheral speaker Sg1 corresponding to the loudspeaker arrival time T11 is obtained. The sounding time can be determined.
[0046]
[Table 1]

[0047]
In step 609, the sum of the lower-order peripheral speakers Sgk reaching the sound receiving point Pg1 before the loudspeaker arrival time T11 from the peripheral speakers Sg1 and the loudspeaker arrival sound pressure levels from the main speaker So is calculated. In this case, at the sound receiving point Pg1, the only loudspeaker sound that arrives before the loudspeaker arrival time from the peripheral speaker Sg1 is the loudspeaker sound from the main speaker So, and therefore the arrival sound pressure of the loudspeaker sound from the main speaker So. In step 610, a sound pressure level difference ΔL1 that gives the preceding sound effect to the loudspeaker arrival sound pressure level Lo1 from the main speaker So is calculated as the level Lo1 as the sum of the arrival sound pressure levels of the preceding loudspeaker sound. This sound pressure level difference ΔL1 is, for example, as shown in FIG. 8, the relationship between the delay time Δt from the preceding sound to the succeeding sound necessary to give the preceding sound effect and the sound pressure level difference ΔL between the preceding sound and the succeeding sound. The equation (ΔL = f (Δt)) is determined in advance by experiments or the like, and the delay time Δt1 required for calculating the sounding time of the peripheral speaker Sg1 is substituted into the relational expression of FIG. 8 (ΔL1 = f (Δt1)). Can be.
[0048]
Based on the obtained sound pressure level difference ΔL1 and the sound pressure reaching sound pressure level Lo1 (= 10logIo1) from the main speaker So, in step 612, the sound sound from the surrounding speaker Sg1 should be indicated at the sound receiving point Pg1 by the sound pressure level L11 ( = 10logIo1 + ΔL1), and by substituting the calculated reached sound pressure level L11 and the distance R11 from the peripheral speaker Sg1 to the sound receiving point Pg1 into the equation (13), the input power W1 of the peripheral speaker Sg1 is determined in step 612. decide.
[0049]
For example, in FIG. 8, when the delay time Δt1 of the arrival time of the peripheral speaker Sg1 is 6 ms, if the difference between the arrival sound pressure level from the peripheral speaker Sg1 and the arrival sound pressure level from the main speaker So is within 6 dB, the sound image Indicates that localization can be obtained. The reached sound pressure level Lo1 at the sound receiving point Pg1 of the main speaker So is reduced due to distance attenuation or the like as shown in Expression (11). Based on the sound pressure level Lo1 at the sound receiving point Pg1 of the main speaker So as a reference, the relative input level of the peripheral speaker Sg1 is set so that the sound pressure level from the peripheral speaker Sg1 at the sound receiving point Pg1 becomes (Lo1 + 6) (dB). The result determined by the attenuator position of the amplifier 12 is -3 (dB) shown in Table 1. As described above, the attenuator position of the amplifier 12 can be adjusted or corrected so that the input power of the peripheral speaker Sgx becomes the value calculated in step 611.
[0050]
After determining the sounding time and input power of the peripheral speaker Sg1 in step 612, the process proceeds to step 613, where it is determined whether the calculation of the sounding time and input power has been completed for all the peripheral speakers. In this case, since the processing of the peripheral speakers Sg2 to Sg4 remains, the process returns to step 605, and the sounding time and input power are determined for the peripheral speaker Sg2 by repeating steps 605 to 612. In the case of the peripheral speaker Sg2, since the lower-order peripheral speaker Sg1 exists, in step 605, in addition to calculating the distance Ro2 from the main speaker So to the sound receiving point Pg2 and the distance R22 from the peripheral speaker Sg2 to the sound receiving point Pg2. , The distance R12 from the lower-order peripheral speaker Sg1 to the sound receiving point Pg2 is calculated. In step 606, the time at which the loud sound from the lower-order peripheral speaker Sg1 reaches the sound receiving point Pg2, and the loud sound reaching sound pressure level L12 are calculated based on the equations (12) and (13).
[0051]
In step 607, a delay time Δt2 that gives a preceding sound effect to the loudspeaker sound arrival time To2 from the main speaker So is obtained. The calculation method is the same as the method described above for the peripheral speaker Sg1. From the sum of the delay time Δt2 and the loud sound arrival time To2 from the main speaker So, a time T22 (= To2 + Δt2) at which the loud sound from the peripheral speaker Sg2 should arrive at the sound receiving point Pg2 is obtained. If the time T22 at which the loud sound should arrive is known, the sounding time of the surrounding speaker Sg2 can be determined in step 608 by substituting the distance R22 from the surrounding speaker Sg2 to the sound receiving point Pg2 into equation (12).
[0052]
In step 609, the loudspeaker arrival time T22 from the peripheral speaker Sg2 and the loudspeaker arrival time T12 from the peripheral speaker Sg1 are compared at the sound receiving point Pg2, and if the loudspeaker arrival time T12 from the peripheral speaker Sg1 precedes. The sum (I12 + Io2) of the squared sound pressure I12 of the loudspeaker arriving from the surrounding speaker Sg1 at the sound receiving point Pg2 and the squared sound pressure Io2 of the arriving loudspeaker sound from the main speaker So is obtained, and the sum (I12 + Io2) ) Into equation (14) to calculate the sum (= 10 log (I12 + Io2)) of the loudspeaker arrival sound pressure level L12 from the surrounding speaker Sg1 and the loudspeaker arrival sound pressure level Lox from the main speaker So. .
[0053]
The inventor has determined that the loud sound from the lower-order peripheral speakers Sgk (k = 1, 2,..., (X−1)) at each sound receiving point Pgx becomes the loud sound from the peripheral speakers Sgx. On the other hand, when the sound arrives first, not only the loudspeaker arrival sound pressure level Lox from the main speaker So, but also the loudspeaker arrival sound pressure level Lkx (= 10logΣIkx) from the lower-order peripheral speaker Sgk and the main speaker So. From the surrounding speaker Sgx above the sound receiving point Pgx so as to give a precedence effect to the sum (= 10 log (ｋIkx + Iox)) with the loud sound reaching sound pressure level Lox (= 10logIox) from the speaker. By determining the pressure level Lxx, it has been found that the leading sound effect can be reliably obtained at all sound receiving points on the wide sound receiving surface Ph. The present invention is based on such findings.
[0054]
In step 610, for example, the sound pressure level difference ΔL2 is obtained by substituting the delay time Δt2 of the sounding time of the peripheral speaker Sg2 into the relational expression of FIG. 8 (ΔL2 = f (Δt2)). By adding the obtained level difference ΔL2 to the sum (= 10log (I12 + Io2)), a sound pressure level L22 (= 10log (I12 + Io2) + ΔL2) at which the sound arrived from the peripheral speaker Sg2 should be shown at the sound receiving point Pg2 is obtained. When the reached sound pressure level is obtained, in step 612, the input power W2 of the peripheral speaker Sg2 can be determined by substituting into the equation (13), as in the case of the peripheral speaker Sg1 described above.
[0055]
After determining the sounding time and input power of the peripheral speaker Sg2 in step 612, the process returns from step 613 to step 605, and repeats steps 605 to 612 for the peripheral speakers Sg3 and Sg4. By repeating steps 605 to 612 in FIG. 5 for all the peripheral speakers Sgx in ascending order, the sounding times and input powers of all the peripheral speakers Sgx can be determined. By inputting an instruction signal of the sounding time and input power of each peripheral speaker Sgx determined by the computer 30 to the signal adjusting device 20, the sounding time and input power of the acoustic signal from the signal transmission device 10 are adjusted for each speaker Sj. Then, the adjusted acoustic signal is input to each speaker Sj via the power amplifier 12 (see FIG. 1).
[0056]
FIG. 9 shows the sound pressure generated at each sound receiving point Sg4 by the loud sounds from the speakers So and Sg1 to Sg4 determined as described above. That is, in this case, the real voice from the speaker O arrives as the earliest preceding sound, the loud sound from the main speaker So closest to the speaker O follows, and then the loud sounds from the peripheral speakers Sg1, Sg2, and Sg3. Reach in order. Finally, the loud sound from the peripheral speaker Sg4 arrives. Since the relationship between the real voice and the loud sound from each speaker and the mutual relation between the loud sounds from the speakers satisfy the condition where the preceding sound effect works, the sound field perception of the listener 2 at the sound receiving point Sg4 is synthesized. It is considered that the subjective level of the sound pressure is as shown by the solid line in FIG. That is, the solid line in FIG. 10 indicates that the loudspeakers from a plurality of distributed speakers are perceived by the listener 2 as one sound.
[0057]
According to the present invention, since the speakers having substantially the same acoustic characteristics are used, there is no difference in the timbre of the speakers, and the difference in the transmission characteristics with respect to both ears of the listener can be kept within an allowable range by appropriately setting the speaker interval. Since the sounding time and input power of the loudspeaker are controlled so that the loud sound from the main loudspeaker becomes the preceding sound, a sound image can be localized in a direction toward the sound source O at any position on the sound receiving surface. Further, since the main speaker can be selected and the peripheral speakers can be ranked according to the movement of the position of the sound source, the sound image can be localized on the moving sound source. It is also possible to localize the sound image to a questioner or the like appearing at an unspecified position.
[0058]
Thus, it is possible to achieve the object of the present invention to provide "a distributed loudspeaker method and a distributed loudspeaker system for localizing a sound image following speaker movement that can change the direction of sound image localization in accordance with the direction of a moving sound source".
[0059]
Preferably, the listening sound pressure level at each sound receiving point Pi is suppressed to a predetermined target sound pressure level Lhc, so that the listener 2 listens at the predetermined target sound pressure level regardless of the sound receiving point Pi on the sound receiving surface Ph. It can be so. In this case, in step 611 of FIG. 5, for example, the listening sound pressure level at the sound receiving point Pg4 reaches the sound pressure level L44 (= 10 log I44) of the loudspeaker reaching the sound from the surrounding speaker Sg4 at the sound receiving point Pg4 in advance. It is determined as the total (= 10logΣ (I44 + ΣIk4 + Io4)) of the sum of the sound pressure levels reached by the lower-order peripheral speakers Sgk (k = 1, 2, 3) and the loudspeaker sound arrival from the main speaker So (= 10log (ΣIk4 + Io4)). When the obtained listening sound pressure level exceeds the target sound pressure level Lhc, the total (= 10 logΣ (I44 + Ik4 + Io4)) of the arrival sound pressure level L44 of the peripheral speaker Pg4 within the range where the preceding sound effect can be obtained is equal to the target sound pressure level. Suppress until it matches Lhc.
[0060]
More preferably, the target sound pressure level Lhc is determined as shown by the solid line in FIG. 11, and control is performed such that the sound pressure level of the synthesized sound field on the sound receiving surface Ph created by each speaker Sj decreases as the distance from the sound source O increases. I do. This attenuation control enables the formation of a sound field that matches the daily experience that the sound becomes smaller when the sound source is away from the sound source O, and acts so that the sound field is perceived by the listener with a natural feeling. Therefore, in a lecture or the like, the listener can listen while feeling that the speaker is talking to the speaker, and the speaker can give psychological feedback such as developing the topic while feeling the audience's reaction well It can be expected that participants will be able to communicate with each other.
[0061]
Although the target sound pressure level Lhc shown in FIG. 11 decreases in accordance with the distance Rh from the sound source O, the slope of the decrease with respect to the distance Rh is smaller than the slope of the sound attenuation in the free sound field. According to the target sound pressure level Lhc, it is possible to increase the sound volume with a sufficiently large sound pressure while giving the listener a natural perception. The black portion of 40 dB or less in the drawing indicates a range in which the sound is too low and hinders the listener's hearing. In the free sound field, even if the speaker utters a little loud voice, it may be difficult for a listener at a distance of about 100 m to hear. However, according to the loudspeaker method using the target sound pressure level Lhc in FIG. It is possible to hear the speaker's utterance with sufficient sound pressure even at a point 100 m away while giving a natural sound field feeling.
[0062]
The inventor sets the maximum sound pressure level Loo from the main speaker So at the sound receiving point Po below the main speaker So as the target sound pressure level Lhc as shown in FIG. By using the inclined target sound pressure level Lhc = Loo-3.0 × logRh whose level decreases in proportion to the logarithm (logRh) of the distance Rh, the listener feels an unnatural shift between the sound and the distance. It has been experimentally confirmed that a sound field can be formed that can hear clear and refreshing natural sound as if the speaker is near even at a position distant from the speaker. However, the proportionality constant of the attenuation of the inclined target sound pressure level Lhc is not limited to the illustrated example.
[0063]
【Example】
FIG. 12 shows an embodiment of the present invention using a composite amplifier 20b in which the signal adjusting device 20 and the amplifier 24 are integrated, and FIG. 13 shows a block diagram of the composite amplifier 20b used in the embodiment. In the illustrated example, the composite amplifiers 20b are connected in tandem with each other, and an audio signal and a control signal are supplied to each of the composite amplifiers 20b by one digital audio signal line 27 and a control signal line 28. The composite amplifier 20b has two sound delay units 21L and 21R and two volume control units 22L and 22R, and can simultaneously process two-channel digital sound signals.
[0064]
The sounding time (delay time) and input power (volume) of the main speaker So and each peripheral speaker Sgx calculated by the computer 30 (see FIG. 14) are sent to the CPU 25 of the composite amplifier 20b via the interface 14 and the control line 28. . The CPU 25 is connected to the sound delay units 21L and 21R and the volume control units 22L and 22R, and adjusts the sounding time and input power of the two-channel digital sound signal in accordance with the processing pattern stored in the pattern memory 26. The adjusted digital audio signal is mixed by the mixer 23 and sent to the speaker via the amplifier 24 and the output line 29. With the use of the composite amplifier 20b in the figure, sounds from two sound sources can be simultaneously processed, and sound images directed to two sound sources on the sound receiving surface Ph can be simultaneously localized.
[0065]
FIG. 14 shows still another embodiment of the present invention using the composite amplifier 20b. Fourteen speakers Sj were dispersedly arranged in a grid pattern at a height of 3.5 m from the floor and at an interval of 5 m × 3.5 m, and a composite amplifier 20 was attached to each speaker Sj. As shown in the figure, the sounds of the first sound source below the speaker S1 and the second sound source below the speaker S4 were collected by the microphones 3 and 4, respectively, and input to the interface 14 via the mixer 11. On the other hand, the computer 30 calculates control signals necessary for sound image localization to the two sound sources according to the flowchart of FIG. The digital audio signal from the mixer 11 and the control signal from the computer 30 were transmitted to the composite amplifier 20 of each speaker Sj via the interface 14. Each composite amplifier 20 was connected in tandem as shown in FIG. In the composite amplifier 20b, the digital audio signal from the microphone 3 sent to the right channel and the digital audio signal from the microphone 4 sent to the left channel are simultaneously processed and sent to each speaker Sj via the mixer 23 and the main amplifier 24. Output.
[0066]
Table 2 shows an example of the control signals used in the embodiment of FIG. The present inventor has confirmed that the sound image heading to the first sound source and the sound image heading to the second sound source can be simultaneously localized on the sound receiving surface Ph by the control signals in Table 2. As can be seen from the embodiment shown in FIG. 14 and Table 2, even when two sound sources O are present in the sound field, such as a speaker on the stage and an interlocutor in the hall, the method of the present invention can be applied. For example, it is possible to perform sound image localization using the speaker as the first sound source and simultaneously perform sound image localization using the interlocutor as the second sound source. By simultaneously performing the two sound image localizations, the speaker and the questioner can hear each other's voices well, so it is easy to talk, and even wherever in the venue the interaction between the two can be heard with a natural feeling You can create a place.
[0067]
[Table 2]

[0068]
【The invention's effect】
As described above in detail, according to the distributed loudspeaker method and the distributed loudspeaker system for localizing a sound image following speaker movement according to the present invention, a plurality of speakers Sj having substantially the same acoustic characteristics are covered above a sound receiving surface. A part of the area is partially superimposed and dispersed downward, the position of the sound source O moving in the sound field where the speakers Sj are dispersed is detected, and the speaker closest to the detected position of the sound source O among the speakers Sj is detected. Is the main speaker So, and the remaining speakers are the peripheral speakers Sgx ranked in ascending order of the distance from the main speaker So, and the sounding time and input of each peripheral speaker are set so that the loudspeaker sound from the main speaker gives a preceding sound effect. Since the level is controlled and the localization direction of the sound image on the sound receiving surface is set to the direction toward the sound source O, it is possible to cope with a change in the position of the sound source. That is, even when the sound source moves, the sound image can be localized in the direction toward the sound source according to the movement by selecting the main speaker again. Therefore, even when the position of the speaker changes, for example, when the speaker moves in a large venue or when the stage position changes, the main speaker So is set substantially within the permissible arrangement range in accordance with the position of the speaker. A state equivalent to the movement is obtained. This makes it possible to follow the direction of sound image localization according to the movement of the sound source position or the appearance of the interrogator at an unspecified position, and in a large lecture hall, the position of the moving speaker or the position of the interrogator can be changed. Problems such as not knowing can be solved.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram of an embodiment according to the present invention.
FIG. 2 is an explanatory diagram of a connection example of an intelligent power amplifier and a tag reading device.
FIG. 3 is a diagram illustrating an example of acoustic characteristics of a speaker.
FIG. 4 is an explanatory diagram of an allowable arrangement range of a preceding sound speaker for obtaining a preceding sound effect.
FIG. 5 is an example of a flowchart of a method of calculating a sounding time and input power of each speaker.
FIG. 6 is an explanatory diagram showing the principle of sound image localization according to the present invention.
FIG. 7 is an explanatory diagram of a preceding sound effect.
FIG. 8 is an explanatory diagram of a relationship between a preceding sound at which a preceding sound effect is obtained and a reinforcing sound.
FIG. 9 is an explanatory diagram of a sound reached from each speaker at the sound receiving point in FIG. 6;
10 is an explanatory diagram of the sound intensity at the sound receiving point in FIG.
FIG. 11 is an explanatory diagram of a target sound pressure level at a sound receiving point.
FIG. 12 is an explanatory diagram of an embodiment of the present invention using a two-channel composite amplifier.
FIG. 13 is an explanatory diagram of a two-channel composite amplifier.
FIG. 14 is an explanatory diagram of another embodiment of the present invention.
FIG. 15 is an explanatory diagram of a conventional centralized loudspeaker system, a distributed loudspeaker system, and a semi-variable loudspeaker system.
FIG. 16 is an explanatory diagram of a reaching sound level at a sound receiving point in the conventional concentrated loudspeaker system, distributed loudspeaker system, and semi-dispersive loudspeaker system.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 3 ... Microphone, 10 ... Signal transmission device, 20 ... Signal adjustment device, 31 ... Speaker selection means, 32 ... Speaker ranking means, 33 ... Main speaker sound instruction means, 34 ... Peripheral speaker sound instruction means, 37 ... Microphone position detection Apparatus, 101: infrared transmitting tag, 103: tag reader, 110: sound source position identification means, O: sound source, Ph: sound receiving surface, Pi: sound receiving point, Sj: speaker, So: main speaker, Sgx: peripheral speaker , Pgx: Lower sound receiving point

Claims (5)

At a predetermined position above the sound receiving surface Ph, a plurality of speakers Sj (1 ≦ j ≦ n) having substantially the same acoustic characteristics used for loudspeaking of the sound of the sound source O are dispersed and arranged downward by partially overlapping the cover areas of the adjacent speakers. A portion on the sound receiving surface Ph that is vertically below each of the speakers Sj is defined as a sound receiving point Pi (1 ≦ i ≦ n), and a sound source O moving in a sound field in which the speakers Sj are dispersedly arranged. Of the speakers Sj, a speaker closest to the position of the detected sound source O is selected as a main speaker So, and the remaining speakers are ranked in ascending order of distance from the main speaker So. Sgx (x = 1, 2,..., (N-1)), and a loud sound of the sound of the sound source O is generated from the main speaker So at time zero with no time delay at a required input power. , The peripheral speaker S For each of x, the sound image is generated at the required input power with a time delay in the ascending order of the order, so that the localization direction of the sound image at each of the sound receiving points Pi follows the movement of the speaker toward the sound source O. A distributed loudspeaker localization method. A distributed loudspeaker method for localizing a sound image following speaker movement according to claim 1,
In each of the peripheral speakers Sgx, in the ascending order of the rank, the loud sound arrival time Txx from the peripheral speaker Sgx at the sound receiving point Pgx below the peripheral speaker Sgx precedes the loud sound arrival time Tox from the main speaker So. The sound generation time of the peripheral speaker Sgx is calculated such that the sound effect is delayed by a delay time Δtx for giving a sound effect (Txx = Tox + Δtx), and the lower order of reaching the sound receiving point Pgx before the sound arrival time Txx from the peripheral speaker Sgx is reached. The sum of the reached sound pressure levels of the surrounding speakers Sgk (k = 1, 2,..., (X-1)) of the rank and the loudspeaker sound from the main speaker So is calculated from each of the speakers Sgk and So. The distance to the sound receiving point Pgx is determined as a function of the input power and the acoustic characteristics of each of the speakers Sgk and So. The sound pressure level Lxx of the loudspeaker sound from the peripheral speaker Sgx at gx is determined to be higher than the sum obtained by the sound pressure level difference ΔLx that gives the preceding sound effect, and corresponds to the determined sound pressure level Lxx. The input power of the peripheral speaker Sgx is calculated, and the loud sound of the sound of the sound source O is sounded from each of the peripheral speakers Sgx at the calculated sounding time at the calculated sound power by following the speaker movement. A distributed loudspeaker method for sound image localization. Loudspeakers Sj (1 ≦ j ≦ n) having substantially the same acoustic characteristics in which the cover areas of the adjacent speakers are partially overlapped and partially downwardly arranged at predetermined positions above the sound receiving surface Ph, and the speakers Sj are dispersedly arranged. Microphone position detection device that detects the position of a sound source O moving in a given sound field and sends out a detection signal, a signal transmission device that transmits an acoustic signal from a sound source O to each of the speakers Sj, the signal transmission device, and each of the above. A signal adjusting device provided between the microphone and the microphone position detecting device, wherein the signal adjusting device is provided between the speaker Sj and adjusts a sounding time and an input power of the sound signal for each of the speakers Sj in accordance with an instruction input and outputs the signals to the respective speakers Sj. Speaker selection means for receiving, as the main speaker So, a speaker closest to the sound source O among the respective speakers Sj, Speaker ranking means for ranking external speakers Sj as peripheral speakers Sgx (x = 1, 2,..., (N-1)) in ascending order of distance from the main speaker So, and the main speaker Main speaker sound instructing means for instructing the sounding time and input power of So, and peripheral speaker sound instructing means for instructing the sounding time and input power of the peripheral speaker Sgx; A distributed loudspeaker system in which an instruction of a sounding time and an input power is input to the signal adjusting device so that a sound image localization direction at each of the sound receiving points Pi follows a speaker movement as a direction toward the sound source O. . 4. The distributed loudspeaker system according to claim 3, wherein the sound image is localized following the movement of the speaker.
The peripheral speaker sound instructing means determines a sound receiving point Pi (1 ≦ i ≦ n) on the sound receiving surface Ph vertically below each of the speakers Sj, and for each of the peripheral speakers Sgx, At the sound receiving point Pgx below the peripheral speaker Sgx, the loudspeaker arrival time Txx from the peripheral speaker Sgx lags behind the loudspeaker arrival time Tox from the main speaker So by a delay time Δtx that gives a preceding sound effect (Txx = Tox + Δtx). ), The sound generation time of the peripheral speaker Sgx is calculated, and the lower-order peripheral speakers Sgk (k = 1, 2,...) That reach the sound receiving point Pgx earlier than the loudspeaker sound arrival time Txx from the peripheral speaker Sgx. , (X-1)) and the sum of the reached sound pressure levels of the loudspeakers from the main speaker So are calculated from the respective speakers Sgk and So. The distance to the sound receiving point Pgx and the input power and acoustic characteristics of each of the speakers Sgk and So were obtained as a function, and the sound pressure level Lxx at which the sound from the surrounding speakers Sgx was reached at the sound receiving point Pgx was obtained. The input power of the peripheral loudspeaker Sgx corresponding to the determined ultimate sound pressure level Lxx is calculated to be higher than the sum by the sound pressure level difference ΔLx that gives a preceding sound effect, and the calculated sounding time and input power are calculated. A distributed loudspeaker system that localizes the sound image following the movement of the speaker. The distributed loudspeaker system according to claim 3 or 4, wherein the sound image is localized following the movement of the speaker.
The microphone position detection device is provided with a microphone, and transmits an tag identification number. The infrared transmission tag is provided on each of the speakers Sj and has a reader identification number and has an infrared transmission tag that has entered the cover area of each of the speakers Sj. A tag reader that reads the tag identification number and sends out the tag identification number and the reader identification number, and detects an arbitrary position of the microphone that moves in the sound field based on the tag identification number and the reader identification number. A distributed loudspeaker system comprising sound source position identification means for localizing a sound image following speaker movement.

Images