JP3772430B2 - Performance data editing device - Google Patents

Description

【００４２】
ただし、この式（１）において、ｍａｘおよびｍｉｎは、チャネル番号ＣＨに対応したＭＩＤＩチャネルにおけるベロシティデータの最大値および最小値であり、既に図１１を参照して説明した処理により求められ、ＲＡＭ１４内に格納されていた情報である。
【００４３】
次に、チャネル番号ＣＨのＭＩＤＩチャネルに対応した全てのベロシティデータを読み終えたか否かを判断する（ステップＳ４０６）。そして、この判断結果が「ＮＯ」の場合はステップＳ４０４に戻り、後続のベロシティデータについて同様の処理を行う。そして、ステップＳ４０６の判断結果が「ＹＥＳ」となった場合には、チャネル番号ＣＨを増加させた後（ステップＳ４０３）、ステップＳ４０７に進んでチャネル番号ＣＨが「１５」を超えているか否かを判断する。この判断結果が「ＮＯ」の場合はステップＳ４０２に進み、ステップＳ４０２において他に処理対象となっているＭＩＤＩチャネルが見つかった場合には、そのＭＩＤＩチャネルについて上述の処理を行う。そして、すべてのＭＩＤＩチャネルについての処理が終了すると、ステップＳ４０７の判断結果が「ＹＥＳ」となり、チェンジャとしての処理が終了する。
【００４４】
以上の処理により、処理対象となったＭＩＤＩチャネルに対応した全ベロシティデータは、レジスタＭＡＸ（ＣＨ）およびＭＩＮ（ＣＨ）に設定された上限値および下限値によって限られた範囲内に分布範囲が拡大される。図１３は、このベロシティデータの分布範囲の拡大の様子を図示したものである。この図に示すように、最大値ｍａｘから最小値ｍｉｎまでの範囲に分布していた各ベロシティデータは、前掲式（１）の適用により、各々相互の大小関係をそのまま維持した状態で、上限値ＭＡＸ（ＣＨ）および下限値ＭＩＮ（ＣＨ）によって挟まれた広い範囲内にその分布範囲が拡大されるのである。
【００４５】
このように本実施形態によれば、指定したＭＩＤＩチャネルについてベロシティデータの分布範囲を任意に広げることができるので、自動演奏時における特定パートの音量の変化の起伏を任意に大きくし、演奏に表情を持たせることができる。
【００４６】
（３）発音重複化編集処理
発音重複化編集処理においては、処理対象となっている各ＭＩＤＩチャネル（ＤＢＬ（ＣＨ）＝“１”となっているＭＩＤＩチャネル）について、当該ＭＩＤＩチャネルに対応した既存のキーオンイベントに対し、これと重複するキーオンイベントを付加する処理を行う。この重複するキーオンイベントの付加を全く無条件に行った場合には発音チャネルが不足するおそれがある。そこで、既存のキーオンイベントの発生状況を時系列的に把握する分析処理を行った後、各時点において発音チャネルの不足が生じないという条件を満たす限りにおいて、重複したキーオンイベントを付加する処理を行う。前者の分析処理はアナライザの実行により行われ、後者の処理はチェンジャの実行により行われる。
【００４７】
アナライザの分析処理（図８のステップＳ２０４に相当する処理）においては、処理対象たるＭＩＤＩチャネルに対応した各キーオンイベントについて、各々と同一のキーオンイベントを付加した場合に発音チャネルが不足しないか否かを判定する。この処理の概要を図１４を参照して説明する。図１４（ａ）（ｂ）において、ＫＯＮ１，ＫＯＮ２，…等は各々キーオンイベントを表しており、ＫＯＦＦ１，ＫＯＦＦ２，…等はこれらのキーオンイベントによって開始された発音動作を終了させるキーオフイベントを表している。これらの各キーオンイベントおよび各キーオフイベントは、ＲＡＭ１４内に記憶された演奏データの中から抽出されたものであり、図面横方向を時間の進行方向とし、各々の再生順序に従って時系列的に並べられている。
【００４８】
ここで、図１４（ａ）および（ｂ）に示す各状況について、発音チャネル不足の有無を如何にして判定するか、その例を説明する。なお、以下では音源ボードが４個の発音チャネルを有していることを前提とする。また、図示されたキーオンイベントおよびキーオフイベントは、すべて発音重複化編集処理の処理対象となっているＭＩＤＩチャネルに対応したものであるとする。
【００４９】
まず、図１４（ａ）に示す状況においては、各キーオンイベントＫＯＮ１〜ＫＯＮ３によって開始される各発音期間は時間軸上において重複しない。従って、各キーオンイベントＫＯＮ１〜ＫＯＮ３はいずれも各々に重複したキーオンイベントを付加することが可能である。
【００５０】
次に、図１４（ｂ）に示す状況について検討する。この場合、各キーオンイベントＫＯＮ１〜ＫＯＮ５によって開始される各発音期間の中には時間軸上において相互に重複するものがある。そこで、各キーオンイベントについて重複したキーオンイベントを付加可能か否かを検討する。
【００５１】
まず、キーオンイベントＫＯＮ１について検討すると、このキーオンイベントの発生タイミングにおいては他のキーオンイベントによる発音は行われていない。また、キーオンイベントＫＯＮ１による発音がキーオフイベントＫＯＦＦ１によって終了するまでの間にキーオンイベントＫＯＮ２およびＫＯＮ３が生じるが、これらのキーオンイベントに対応した発音を行ったとしても、依然として発音チャネルが１個余る。従って、キーオンイベントＫＯＮ１による発音が行われる期間、さらに１個の発音チャネルを使用して発音を行ったとしても、発音チャネル不足は生じない。よって、キーオンイベントＫＯＮ１については、これと同じキーオンイベントを付加し、同一音について発音チャネルを２個使用した重複発音を行ってもよいという判定結果となる。
【００５２】
次に、キーオンイベントＫＯＮ２について検討する。上記判定の結果に従うとすると、キーオンイベントＫＯＮ１に関しては発音チャネルを２個使用して重複した発音が行われるため、キーオンイベントＫＯＮ２に対応した発音を行う際には発音チャネルを３個使用することとなる。そして、キーオンイベントＫＯＮ２による発音がキーオフイベントＫＯＦＦ２によって終了するまでの間に、キーオンイベントＫＯＮ３に対応した発音が開始される。ここで、キーオンイベントＫＯＮ３に対応した発音開始時点においては、依然としてキーオンイベントＫＯＮ１に対応した発音が行われているので、この時点において発音チャネルを４個使用することとなる。従って、仮にキーオンイベントＫＯＮ２について発音チャネルを２個使用した発音を行ってしまうと、キーオンイベントＫＯＮ３に対応した発音を開始する時点において発音チャネルが不足してしまう。よって、キーオンイベントＫＯＮ２については、発音チャネルを２個使用した重複発音は許可できないという判定結果となる。
【００５３】
次に、キーオンイベントＫＯＮ３について検討する。上記各判定の結果に従うとすると、キーオンイベントＫＯＮ１に関しては発音チャネルを２個使用して重複した発音が行われ、キーオンイベントＫＯＮ２に関しては発音チャネルを１個使用した発音が行われるため、キーオンイベントＫＯＮ３に対応した発音を行う際には発音チャネルを４個使用することとなる。従って、キーオンイベントＫＯＮ３に関しては、発音チャネルを２個使用した重複発音をすることができないのは明らかである。
【００５４】
次に、キーオンイベントＫＯＮ４について検討する。このキーオンイベントＫＯＮ４に対応した発音を開始する時点においては、キーオンイベントＫＯＮ１に対応した発音が既に終了している。従って、この時点においては、キーオンイベントＫＯＮ２、ＫＯＮ３およびＫＯＮ４に対応した発音を行うために３個の発音チャネルを使用することとなり、発音チャネルが１個余る。また、キーオンイベントＫＯＮ４による発音がキーオフイベントＫＯＦＦ４によって終了するまでの間、他のキーオンイベントに対応した発音は開始されない。よって、キーオンイベントＫＯＮ４については、発音チャネルを２個使用した重複発音を認めてよいという判定結果となる。
【００５５】
以上説明した判定は、複数のキーオンイベントに対応した発音期間が時間軸上において重複し、いずれかのキーオンイベントについて重複発音を行うと発音チャネルの不足が生じる場合には、時間的に先に再生されるキーオンイベントを優先して重複発音を許可するというルールに従って行われるものである。なお、このようなルールの他、発音重複化についての各キーオンイベントの競合を回避する各種のルールが考えられるが、その説明は省略する。
【００５６】
発音重複化編集処理の分析処理においては、以上のような原理に従い、重複発音の適用の可否の判定を処理対象たるＭＩＤＩチャネルに対応した各キーオンイベントについて行うのである。
【００５７】
この分析処理を行うためには、各キーオンイベントおよび各キーオフイベントの発生状況を時系列的に捉える必要がある。そこで、本実施形態においては、演奏データに含まれているステップタイムを参照し、各キーオンイベントに対応した発音期間の時間軸上での占有位置を求める。すなわち、演奏データ中の各イベントの前には、図１５に例示するようにステップタイムを表すデータが付加されている。このステップタイムは、その直前のイベントを再生してから当該イベント（ステップタイムの直後のイベント）を再生するまでの待機時間である。従って、任意のイベントについて当該イベントよりも前に存在する全てのステップタイムの合計値を演算すれば、自動演奏の開始時刻を基準とした当該イベントの相対的な再生時刻を求めることができる。本実施形態においては、各キーオンイベントおよび各キーオフイベントについてこのような演算を行うことにより、各キーオンイベントに対応した発音期間の時間軸上での占有位置を求めるのである。
【００５８】
次に図１６を参照し、発音重複化編集処理の分析処理の具体的なフローについて説明する。まず、処理対象となっているＭＩＤＩチャネルに対応した演奏データの中から先頭のイベントを取り込む（ステップＳ５０１）。次に、この取り込んだイベントがキーオンイベントか否かを判断する（ステップＳ５０２）。この判断結果が「ＹＥＳ」である場合にはステップＳ５０３に進み、「ＮＯ」の場合はステップＳ５０８に進む。
【００５９】
次にステップＳ５０３に進むと、今回取り込んだキーオンイベントの周辺イベントの調査を行う。すなわち、上述したようにステップタイムに基づいて把握される各キーオンイベントに対応した各発音期間に着目し、次の事項の調査を行う。
【００６０】
ａ．今回取り込んだキーオンイベントに対応した発音期間が他のキーオンイベントに対応した発音期間と時間軸上において重複していないかどうか。
ｂ．発音期間の重複が認められる場合において、今回取り込んだキーオンイベントに重複発音を適用した場合に発音チャネルの不足が生じないかどうか。
【００６１】
次にステップＳ５０４に進み、上記ステップＳ５０３での調査結果に基づき、今回取り込んだキーオンイベントに対し、重複発音を適用可能か否かを判断する。この判断の結果は、次の３通りに分れる。
【００６２】
ａ．無条件に重複発音を認めて構わない。
この判断がなされる状況を図１７（ａ）に例示する。この例では、キーオンイベントＫＯＮ１およびＫＯＮ２に対応した各発音期間は、他のキーオンイベントに対応した発音期間と全く重複していない。よって、今回取り込んだキーオンイベントがこの例に示すキーオンイベントＫＯＮ１およびＫＯＮ２のような状況に置かれている場合には、無条件に重複発音を適用可能という判定結果となる。
【００６３】
ｂ．条件付きで重複発音を認めてもよい。すなわち、今回取り込んだキーオンイベントと競合する他のキーオンイベントがあるため、今回取り込んだキーオンイベントに重複発音を適用可能か否かはこれらの他のキーオンイベントに重複発音を適用するか否かによる。
【００６４】
この判断がなされる状況を図１７（ｂ）に例示する。この例では、キーオンイベントＫＯＮ１、ＫＯＮ２およびＫＯＮ３は各々の発音期間が時間軸上において重複しているため、これらのうち１個のみについて重複発音を適用することが許される。また、キーオンイベントＫＯＮ４は、キーオンイベントＫＯＮ２およびＫＯＮ３と発音期間が重複している。従って、キーオンイベントＫＯＮ４については、キーオンイベントＫＯＮ２およびＫＯＮ３のいずれについても重複発音が適用されていない場合に限り、重複発音を認めてよい。今回取り込んだキーオンイベントがこの例に示すキーオンイベントＫＯＮ１〜ＫＯＮ４のような状況に置かれている場合には、これらに対し重複発音を適用可能か否かは競合する他のキーオンイベントに対する重複発音の適用如何によることとなる。
【００６５】
ｃ．今回取り込んだキーオンイベントに重複発音を認める余地は全くない。
キーオンイベントの追加を行わなくても既に発音チャネルが全て使用されており、重複発音を行うと必ず発音チャネルが不足してしまう状況である。
【００６６】
ステップＳ５０４の判定において、判定結果が上記ａまたはｂに該当する場合にはステップＳ５０５に進み、上記ｃに該当する場合にはステップＳ５０８に進む。
【００６７】
次にステップＳ５０５に進むと、ステップＳ５０４の判定結果が上記ｂであるか否かを判断する。この判断結果が「ＹＥＳ」の場合はステップＳ５０６に進み、今回取り込んだキーオンイベントの内容と時間軸上での位置を示す情報をＲＡＭ１４内に記録する。そして、ステップＳ５０８に進む。
【００６８】
一方、ステップＳ５０４の判定結果が「ＮＯ」の場合、すなわち、今回取り込んだキーオンイベントが上記の状況ｂに置かれている場合にはステップＳ５０７に進み、競合を回避するための調整を行う。この調整は、例えば既に図１４を参照して説明したように先に再生されるキーオンイベントを優先して重複発音を適用するルールに従って行う。この方法の他、例えば競合しているキーオンイベントをＣＲＴディスプレイ２に表示し、いずれのキーオンイベントに重複発音を適用するかを使用者に選択させるようにしてもよい。重複発音を適用するキーオンイベントを決定すると、そのキーオンイベントの内容および時間軸上での位置を示す情報をＲＡＭ１４内に記録し、ステップＳ５０８に進む。
【００６９】
次にステップＳ５０８に進むと、処理対象となっているＭＩＤＩチャネルに対応した全てのイベントについて以上説明した処理を実行したか否か、すなわち、分析を終了してよいか否かを判断する。この判断結果が「ＮＯ」の場合は、次のイベントを取り込み（ステップＳ５０９）、ステップＳ５０２へ戻る。以下、同様にして処理対象たるＭＩＤＩチャネルに対応した全てのイベントについてステップＳ５０２〜Ｓ５０８からなる処理が実行される。そして、ステップＳ５０８の判断結果が「ＹＥＳ」になることを以て、このアナライザとしての処理を終了する。
【００７０】
チェンジャとしての処理は、図１８に示すフローに従って実行される。まず、ＭＩＤＩチャネルのチャネル番号ＣＨを「０」に初期設定する（ステップＳ６０１）。次いでダブラレジスタＤＢＬ（ＣＨ）の内容が“１”か否か、すなわち、現在のチャネル番号ＣＨに対応したＭＩＤＩチャネルが発音重複化編集処理の処理対象となっているか否かを判断する（ステップＳ６０２）。そして、この判断結果が「ＮＯ」である場合はチャネル番号ＣＨを１だけ増加させた後（ステップＳ６０５）、チャネル番号ＣＨが「１５」を超えているか否かを判定し（ステップＳ６０４）、超えていなければステップＳ６０２の判断を再び行う。
【００７１】
一方、レジスタＤＢＬ（ＣＨ）の内容が“１”となっている場合には、ステップＳ６０３に進んでキーオンイベント挿入処理を実行する。すなわち、チャネル番号ＣＨに対応したＭＩＤＩチャネルのイベントをＲＡＭ１４から１個ずつ順次読み出し、読み出したイベントをＲＡＭ１４内の別のエリアに順次書込んでゆく。その際に、上述したアナライザの実行の際にＲＡＭ１４に格納した各キーオンイベントに関する情報を参照し、ＲＡＭ１４から読み出した各イベントが重複発音を適用することとしたキーオンイベントであるか否かを判定する。そして、重複発音を適用することとしたキーオンイベントについては、これと同一内容のキーオンイベントを付加し、ＲＡＭ１４へ書込む。このキーオンイベント挿入処理が終わると、ステップＳ６０５に進んでチャネル番号ＣＨを１増加させる。そして、ステップＳ６０４に進んで、チャネル番号ＣＨが「１５」を超えたか否かを判断する。この判断結果が「ＮＯ」の場合はステップＳ６０２に戻る。その後、処理が進んでチャネル番号ＣＨが「１５」を超えることにより、ステップＳ６０４の判断結果が「ＹＥＳ」となり、チェンジャとしての処理が終了する。
【００７２】
以上のように、本実施形態によれば、所望のＭＩＤＩチャネルのキーオンイベントに対し重複発音を適用することができるため、所望のパートの音を芯のある太い音とすることができる。
【００７３】
（４）タイミングチューナ編集処理
最初に、図１９を参照し、タイミングチューナ編集処理の概要を説明する。図１９において、横方向の直線は時間軸を示しており、この時間軸の上側に並べられた一連の白抜きの三角マークは各々、このタイミングチューナ編集処理の処理対象たるキーオンイベントの再生タイミングを示している。また、時間軸の下側には一連の黒塗りの三角マークが記されているが、これらは使用者が発音を望んでいるタイミングを示している。
【００７４】
このタイミングチューナ編集処理においては、処理対象たるＭＩＤＩチャネル（ＴＵＮ（ＣＨ）＝“１”となっているＭＩＤＩチャネル）に対応したキーオンイベントまたはキーオフイベントについて、各々が使用者の望むタイミングで再生されるように再生タイミングの補正を行うものである。タイミングチューナ編集処理は、既存のキーオンイベントまたはキーオフイベントの再生タイミングと使用者が望む再生タイミングとのずれ時間、すなわち、図１９に示すΔｔ1，Δｔ2，…を求める分析処理と、この分析処理の結果に基づいて各キーオンイベントおよびキーオフイベントの再生タイミングを補正する処理とにより構成される。前者の分析処理を行うためのプログラムがアナライザであり、後者の処理を行うためのプログラムがチェンジャである。
【００７５】
アナライザを実行するに当たり、使用者が望む再生タイミングを表すデータを予め準備する必要がある。このデータを得るための方法としては次の２通りがある。
【００７６】
ａ．使用者が実演奏をすることによる方法
例えば使用者が他の電子楽器を使用して特定パート（例えばリズムパート）の演奏を行う。その際に得られる実演奏データをＭＩＤＩインタフェース１７からこの演奏データ編集装置に供給し、実演奏データ中のキーオンイベントまたはキーオフイベントの再生タイミングを使用者が望む再生タイミングとして使用する。
【００７７】
ｂ．図２０に示すように使用者がキーボード３のスペースキーを叩くことにより所望の再生タイミングを指定する実演奏データを作成する方法
この方法を利用する場合、使用者は、演奏データ編集装置の動作モードをＭＩＤＩデータ作成モードに設定した後、スペースキーのみを使用したリズム演奏をする。すなわち、発音を開始させたいタイミングにおいてスペースキーをオンとし、発音を終了させたいタイミングにおいてスペースキーをオフとすればよい。従って、この方法は、電子楽器演奏に不慣れな使用者にとっては便利である。
【００７８】
図２１は、この方法に従ってリズム系の楽音のキーオンイベントの再生タイミングを表す実演奏データを作成する実演奏データ作成プログラムのフローを示すものである。以下、このプログラムの処理内容を説明する。
【００７９】
このプログラムは所定のコマンドが入力されることにより開始される。まず、ステップＳ７０１に進むと、スペースキーのオンイベントがあったか否かを判断する。この判断結果が「ＮＯ」の場合にはステップＳ７０４に進み、処理終了の指示がなされたか否か（図示しない終了ボタンが押されたか否か）を判断する。そして、この判断結果が「ＮＯ」の場合はステップＳ７０１に戻る。
【００８０】
スペースキーのオンイベントが発生すると、ステップＳ７０１の判断結果が「ＹＥＳ」となってステップＳ７０２に進む。そして、現在の動作モードがＭＩＤＩデータ作成モードか否かを判断する。この判断結果が「ＮＯ」の場合は何等のデータを作成することもなくこのプログラムを終了する。
【００８１】
一方、ステップＳ７０２の判断結果が「ＹＥＳ」の場合はステップＳ７０３に進み、ＭＩＤＩエンコード処理を実行する。すなわち、このプログラムの実行を開始してからの経過時間に相当するステップタイムと所定のキーオンイベントを作成し、ＲＡＭ１４内の所定のエリアに書込む。そして、この書込みを終えた後、ステップＳ７０１に戻る。
【００８２】
その後、再びスペースキーのオンイベントが発生すると、ステップＳ７０１，Ｓ７０２を介してステップＳ７０３に進み、再びＭＩＤＩエンコード処理を実行することとなる。この場合には、前回、ステップタイムおよびキーオンイベントの書込みを行ってから現在に至るまでの経過時間に相当するステップタイムと所定のキーオンイベントを作成し、ＲＡＭ１４内の所定のエリアに書込む。以後、同様に、スペースキーのオンイベントが検知される毎に、ステップタイムとキーオンイベントのＲＡＭ１４への記録を行ってゆく。そして、スペースキーの操作が途絶え、処理終了の指示がなされるとステップＳ７０４の判断結果が「ＹＥＳ」となり、処理が終了する。
【００８３】
以上の処理により、使用者が望むキーオンイベントの再生タイミングを表す実演奏データがＲＡＭ１４内に得られる。
【００８４】
なお、持続系の音のキーオンイベントおよびキーオフイベントの両方の再生タイミングを指定するデータをスペースキー操作によって作成したい場合には、図２１に示すプログラムに対し、スペースキーのオンイベントだけでなくオフイベントも判定するように処理を追加すればよい。
【００８５】
このようにして実演奏データが得られると、次にアナライザが実行され、タイミングチューナ編集処理の処理対象となっているＭＩＤＩチャネル（ＴＵＮ（ＣＨ）＝“１”となっているＭＩＤＩチャネル）について、図２２に示す分析処理が実行される。まず、ステップＳ８０１に進むと、ＲＡＭ１４内の演奏データからキーオンイベント以外のイベントの除去された楽譜データを作成する。すなわち、処理対象たるＭＩＤＩチャネルに対応した一連のイベントとこれらの各イベント間に介挿されたステップタイムをＲＡＭ１４から順次読み出してゆく。そして、この読み出しを行うのと並行し、次の処理を行う。
【００８６】
ａ．キーオンイベントが読み出されるまでの間は、その間に読み出されるステップタイムを累算してゆく。
ｂ．キーオンイベントが読み出された時点でステップタイムの累算値と当該キーオンイベントをＲＡＭ１４内の楽譜データ用の記憶エリアに書込む。
ｃ．この書込みを行った後は、ステップタイムの累算値を０に戻し、ａの処理を行う。
【００８７】
以上の処理が行われる結果、処理対象たるＭＩＤＩチャネルに対応したイベントが全て読み出された時点で、ＲＡＭ１４内にキーオンイベント以外のイベントを含まない楽譜データが得られる。
【００８８】
次にステップＳ８０２に進み、ＲＡＭ１４内の楽譜データおよび実演奏データの各々からイベントおよびその前に置かれたステップタイムを順次取り出してゆく。次いでステップＳ８０３に進み、このようにして取り出された楽譜データの各ステップタイムと実演奏データの各ステップタイムとを各々対応するもの同士について比較し、差分（すなわち、前掲図１９に示すΔｔ1，Δｔ2，…等のずれ時間）を求める。そして、この差分を各キーオンイベントに対応させてＲＡＭ１４内に書込み、分析処理を終了する。
【００８９】
チェンジャとしての処理は、図２３に示すフローに従って実行される。まず、ＭＩＤＩチャネルのチャネル番号ＣＨを「０」に初期設定する（ステップＳ９０１）。次いでチューナレジスタＴＵＮ（ＣＨ）の内容が“１”か否か、すなわち、現在のチャネル番号ＣＨに対応したＭＩＤＩチャネルがタイミングチューナ編集処理の処理対象となっているか否かを判断する（ステップＳ９０２）。そして、この判断結果が「ＮＯ」である場合はチャネル番号ＣＨを１だけ増加させ（ステップＳ９０５）、ステップＳ９０４に進んでチャネル番号ＣＨが「１５」を超えたか否かが判断される。この判断が「ＮＯ」の場合は、再びステップＳ９０２の処理を行う。
【００９０】
一方、レジスタＴＵＮ（ＣＨ）の内容が“１”となっている場合には、ステップＳ９０３に進んでステップタイム変更処理を実行する。すなわち、チャネル番号ＣＨに対応したＭＩＤＩチャネルのイベントをＲＡＭ１４から１個ずつ順次読み出し、読み出したイベントをＲＡＭ１４内の別のエリアに順次書込んでゆく。この間、キーオンイベントが読み出された場合には、ＲＡＭ１４に格納された当該キーオンイベントに対応したステップタイムの差分を参照する。そして、当該キーオンイベントの直前のステップタイムについては、この差分に相当するだけ値の補正を行い、この補正後のステップタイムを当該キーオンイベントと共にＲＡＭ１４に書込む。この処理が行われる結果、処理対象たるＭＩＤＩチャネルに対応した各キーオンイベントの再生タイミングは、実演奏データの各キーオンイベントの再生タイミングと一致するように補正される。なお、上記差分をそのまま補正に使用するのではなく、１より小さい所定の係数を差分に乗じたものを補正に使用し、キーオンイベントの再生タイミングを実演奏データのキーオンタイミングに近づけるようにしてもよい。
【００９１】
このステップタイム変更処理が終わるとステップＳ９０５に進み、チャネル番号ＣＨを増加させる。そして、ステップＳ９０４に進み、チャネル番号ＣＨが「１５」を超えたか否かを判断する。この判断結果が「ＮＯ」の場合は、ステップＳ９０２に戻る。その後、処理が進んでチャネル番号ＣＨが「１５」を超えると、ステップＳ９０４の判断結果が「ＹＥＳ」となり、チェンジャとしての処理が終了する。
【００９２】
以上のように、本実施形態によれば、上記タイミングチューナ編集処理により、特定のパートに対応した楽音の再生タイミングを使用者の望むタイミングに微調整することができるため、自動演奏に使用者の好みの乗りを持たせることができる。
【００９３】
【発明の効果】
以上説明したように、この発明によれば、自動演奏の際、本来発音されるべき他の楽音が発音チャネル数不足により発音されなくなるという弊害を招くことなく、所望のパートについて芯のある太い音が再生されるように演奏データを編集することができる。
【図面の簡単な説明】
【図１】 この発明の一実施形態である演奏データ編集装置の外観を示す図である。
【図２】 同実施形態の電気的構成を示すブロック図である。
【図３】 同実施形態の制御用レジスタを示す図である。
【図４】 同実施形態のＲＯＭ１５内の制御プログラムを示す図である。
【図５】 同実施形態の動作を示すフローチャートである。
【図６】 同実施形態の動作を示すフローチャートである。
【図７】 同実施形態の動作を示すフローチャートである。
【図８】 同実施形態の動作を示すフローチャートである。
【図９】 同実施形態のベロシティエキスパンダ編集処理を説明する図である。
【図１０】 ベロシティエキスパンダ編集処理の処理対象であるキーオンイベントを示す図である。
【図１１】 同実施形態の動作を示すフローチャートである。
【図１２】 同実施形態の動作を示すフローチャートである。
【図１３】 ベロシティエキスパンダ編集処理の効果を示す図である。
【図１４】 同実施形態の発音重複化編集処理を説明する図である。
【図１５】 発音重複化編集処理の処理対象を示す図である。
【図１６】 同実施形態の動作を示すフローチャートである。
【図１７】 発音重複化編集処理の分析処理を説明する図である。
【図１８】 同実施形態の動作を示すフローチャートである。
【図１９】 同実施形態のタイミングチューナ編集処理を説明する図である。
【図２０】 同実施形態における実演奏データの作成方法を説明する図である。
【図２１】 同実施形態の動作を示すフローチャートである。
【図２２】 同実施形態の動作を示すフローチャートである。
【図２３】 同実施形態の動作を示すフローチャートである。
【符号の説明】
１１……ＣＰＵ（演奏データ付加手段）、
１２……表示部、１３……操作部（パート指定手段）、
１４……ＲＡＭ（演奏データ記憶手段）、
１５……ＲＯＭ、１６……ハードディスク装置、
１７……ＭＩＤＩインタフェース、
６………音源ボード、５Ｌ，５Ｒ……スピーカ。[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a performance data editing apparatus for editing performance data.
[0002]
[Prior art]
Recently, various personal computers equipped with a sound source and software for handling performance data have been provided so that anyone can easily enjoy computer music. That is, according to the personal computer corresponding to this type of computer music, the user can edit the series of musical performance data by inputting performance data step by step by operating a keyboard or the like. Therefore, even a person who is not accustomed to playing a musical instrument or who cannot play a musical instrument can enjoy musical compositions such as composition and arrangement.
[0003]
[Problems to be solved by the invention]
Now, many sound sources such as those mounted on personal computers can only produce so-called thin sounds. On the other hand, in an electronic musical instrument such as a synthesizer, a technique for generating a thick core sound by simultaneously generating the same musical sound using a plurality of sound generation channels is used. Even in automatic performance using a computer, there is a strong demand for playing a specific part (for example, a melody part) with a thick sound, and application of the above technique is desired. One method for meeting this requirement is, for example, a method of duplicating and adding to the performance data the same key-on event for all key-on events in the performance data corresponding to a specific part. However, when such a simple method is adopted, the sound generation corresponding to the newly added key-on event is performed, so that the number of sound generation channels of the sound source is insufficient, and other musical sounds that should be originally generated cannot be generated. This may happen.
[0004]
The present invention has been made under the background described above, and solves such a problem, and a performance data editing apparatus capable of editing performance data so that a thick core sound is reproduced for a desired part. The purpose is to provide.
[0005]
[Means for Solving the Problems]
  This inventionA performance data editing device for editing performance data for each part to be generated,
  SaidIn performance dataMultipleofPart of1 from the insideSpecify a partFingerMeans,
  Analyzing the performance data, in the part designated by the designation meansInvestigate the pronunciation period of each tone,Within the pronunciation period of one musical tone in the part,at the same timeWithin the part that is being pronouncedMusicalFind the maximum number, the maximum numberIs a predetermined numberIf less than,For generating a musical tone having the same pronunciation period as that of the first musical tonedataTo the performance dataAdditionDoWith additional means
  The gist of the present invention is a performance data editing apparatus comprising:
[0006]
  Further, the invention according to claim 2 is the performance data editing apparatus according to claim 1, wherein when the maximum number is less than a predetermined number, the adding means outputs the same musical sound as the first musical sound. The gist is to add to the performance data data for sound generation in the same sound generation period as one musical sound.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments will be described below for easier understanding of the present invention.
Such an embodiment shows one aspect of the present invention, and is not intended to limit the present invention, and can be arbitrarily changed within the scope of the present invention.
[0008]
A. Constitution
FIG. 1 is a view showing an appearance of a performance data editing apparatus according to an embodiment of the present invention. In this embodiment, software for editing performance data is installed in a general personal computer with a built-in sound source. Each component will be described with reference to FIG. 1. 1 is a computer main body, 2 is a CRT display, 3 is a keyboard, 4 is a mouse, 5L and 5R are speakers. Reference numeral 6 denotes a multi-tone sound source board that can simultaneously generate a plurality of sounds through a plurality of sound generation channels, and is accommodated in the computer main body 1. In the example shown in FIG. 1, a keyboard electronic musical instrument 8 is connected to the computer main body 1 via a MIDI cable 7.
[0009]
An internal electrical configuration of the computer main body 1 is shown in FIG. In FIG. 2, 11 is a CPU for controlling each part of the performance data editing apparatus. Reference numeral 12 denotes a display unit which controls display of display means such as the CRT display 2 under the control of the CPU 11. Reference numeral 13 denotes an operation unit, and the CPU 11 detects an operation performed on the operation means such as the keyboard 3 and the mouse 4 via the operation unit 13. Reference numeral 14 denotes a RAM provided as a work area for the CPU 11, in which control registers necessary for the CPU 11 to perform various controls are set. FIG. 3 shows the main ones of these control registers. The RAM 14 is used as a storage means for performance data during automatic performance or editing of performance data. A ROM 15 stores various control programs executed by the CPU 11 in addition to an editing program for editing performance data, as shown in FIG.
[0010]
Reference numeral 16 denotes a hard disk device which is a large-capacity storage means provided for storing various information in addition to performance data. Reference numeral 17 denotes a MIDI interface, and the CPU 11 exchanges MIDI data with an external electronic musical instrument or the like via the MIDI interface 17. The MIDI cable 7 in FIG. 1 is connected to the MIDI interface 17. Further, a musical sound generating means including the sound source board 6 and the speakers 5L and 5R shown in FIG. 1 is also placed under the control of the CPU 11.
[0011]
There are the following four types of operation modes of the performance data editing apparatus. Any of the operation modes is provided when the CPU 11 executes a control program corresponding to each mode stored in the ROM 15.
[0012]
(1) Memory mode
When the storage mode is set, the CPU 11 executes the storage program in the ROM 15. In this storage mode, performance data sequentially input by step input in the MIDI data creation mode described later or performance data input from an external electronic musical instrument or the like via the MIDI interface 17 is stored in the RAM 14.
[0013]
(2) Playback mode
When the reproduction mode is set, the CPU 11 executes the reproduction program in the ROM 15. In this playback mode, under the control of the CPU 11, various control information is sent to the tone generator board 6 according to the performance data stored in the RAM 14, and automatic performance based on the performance data is performed.
[0014]
(3) Edit mode
When the editing mode is set, the CPU 11 executes the editing program in the ROM 15. In this editing mode, performance data can be changed, deleted, and added, and the following characteristic editing processing can be executed. In any case, when the user designates a specific MIDI channel, only the performance data corresponding to the specific MIDI channel among all performance data is subjected to the editing process.
[0015]
(A) Velocity expander editing process
One component of the performance data is velocity data (volume data) that determines the volume at the time of pronunciation. In this velocity expander editing process, velocity data corresponding to a specific MIDI channel is processed, and the distribution range of each velocity data is expanded so that the entire velocity data to be processed falls within a predetermined range. . That is, the velocity expander editing process has the effect of increasing the undulation of the change in volume of a specific part during automatic performance and making the specific part stand out.
[0016]
(B) Overlapping pronunciation editing process
When the same musical tone is formed by overlapping a plurality of tone generation channels of the tone generator board 6, a thick tone with a core can be obtained as compared with the case where one tone generation channel is used. The pronunciation duplication editing process is a process performed to obtain this effect for a specific part. In other words, the pronunciation duplication editing process is for processing a key-on event corresponding to a specific MIDI channel, and adds a key-on event that overlaps the existing key-on event. Here, if the addition of the duplicate key-on event is performed unconditionally, there is a possibility that the sound generation channel is insufficient. Therefore, in the present embodiment, the occurrence state of an existing key-on event is grasped in a time series, and duplicate key-on events are added as long as the condition that a sound channel does not run out at each time point is satisfied.
[0017]
(C) Timing tuner editing process
The timing tuner editing process is an editing process for adjusting the playback timing of a key-on event corresponding to a specific MIDI channel based on actual performance data that records actual musical instrument performance or actual performance data created by operating the keyboard 3. .
[0018]
(4) MIDI data creation mode
This is an operation mode in which operation data such as a keyboard 3 and a mouse 4 are operated to sequentially create performance data step by step.
[0019]
Next, with reference to FIG. 3, the main ones of the control registers set in the RAM 14 will be described.
[0020]
(A) Mode register
The user can designate a desired one of the storage mode, the playback mode, the edit mode, and the MIDI data creation mode by operating the keyboard 3 or the mouse 4. Mode designation information “0” to “3” is assigned to each operation mode, and the mode designation information corresponding to the operation mode designated by the user is set in this mode register.
[0021]
(B) Expander register EXP, MIN, MAX
This expander register is means for storing information for controlling the above-described velocity expander editing process, and is constituted by three types of registers EXP, MAX, and MIN. The register EXP (CH) (CH = 0 to 15) is a register for storing information indicating whether or not to apply the velocity expander editing process for each MIDI channel. When applying the panda editing process, EXP (CH) = “1”, and when not applying, EXP (CH) = “0”. Registers MAX (CH) (CH = 0 to 15) and MIN (CH) (CH = 0 to 15) indicate the upper limit value and lower limit value of the distribution range of velocity data after execution of the velocity expander editing process for each MIDI channel. Is a register for storing each. The upper limit value and the lower limit value of the velocity data corresponding to each MIDI channel can be set by the user via the operation means such as the keyboard 3, but if there is no setting by the user, a default value is set.
[0022]
(C) Doubler register DBL
The doubler register DBL (CH) (CH = 0 to 15) is a register for storing information indicating whether or not the above-described pronunciation duplication editing process is applied for each MIDI channel. On the other hand, DBL (CH) = “1” is set when the pronunciation overlapping editing process is applied, and DBL (CH) = “0” is set when not applied.
[0023]
(D) Tuner register TUN
The tuner register TUN (CH) (CH = 0 to 15) is a register that stores information indicating whether or not to apply the timing tuner editing process described above for each MIDI channel. When the timing tuner editing process is applied, TUN (CH) = “1”, and when not applied, TUN (CH) = “0”.
[0024]
B. Action
The operation of this embodiment will be described below.
(1) General operation
When operating means such as the keyboard 3 and mouse 4 are operated, an interrupt request is made from the operation unit 13 to the CPU 11. As a result, the CPU 11 executes the operation detection program (FIG. 4) stored in the ROM 15 as an interrupt process.
[0025]
The flow of this operation detection program is as shown in FIG. First, proceeding to step S101, the CPU 11 detects the content of the operation via the operation unit 13, and determines whether or not the operation is an operation requesting mode setting. If the determination result is “YES”, the process proceeds to step S106, and the operation requested by the operation among the mode designation information (“0” to “3”) of the storage mode, the reproduction mode, the edit mode, and the MIDI data creation mode. Information corresponding to the mode is set in the mode register, and the operation detection program is terminated.
[0026]
If the determination result in step S101 is “NO”, the process proceeds to step S102, and it is determined whether or not the operation detected via the operation unit 3 is an operation instructing the velocity expander editing process. If this determination is “YES”, the flow proceeds to step S107, and control information necessary for executing the velocity expander editing process is set. That is, in the operation for instructing the velocity expander editing process, the specification of the MIDI channel number to be processed, the upper limit value and the lower limit value of the velocity data distribution range after the velocity expander editing process are input. Then, “1” is set in the register EXP (CH) corresponding to the MIDI channel number CH, and the upper limit value and the lower limit value are set in the registers MAX (CH) and MIN (CH). The user can omit the specification of the upper limit value and the lower limit value. In this case, predetermined default values (for example, “120” and “10”) are set for the registers MAX (CH) and MIN (CH). Set. When the data setting for each register is completed, the operation detection program is terminated.
[0027]
If the determination result in step S102 is “NO”, the process proceeds to step S103, and it is determined whether or not the operation detected via the operation unit 3 is an operation instructing the pronunciation overlapping editing process. If the determination result is “YES”, the process proceeds to step S108, where the number of the MIDI channel designated as the processing target of the pronunciation duplication editing process is detected via the operation unit 13, and the MIDI channel number CH is set. “1” is set in the corresponding doubler register DBL (CH), and the operation detection program is terminated.
[0028]
If the determination result in step S103 is “NO”, the process proceeds to step S104, and it is determined whether or not the operation detected via the operation unit 3 is an operation instructing timing tuner editing processing. If the determination result is “YES”, the process proceeds to step S109, the number of the MIDI channel designated as the processing target of the timing tuner editing process is detected via the operation unit 13, and the number corresponds to the MIDI channel number CH. The tuner register TUN (CH) is set to “1”, and the operation detection program is terminated.
[0029]
If the determination result in step S104 is “NO”, an operation detection process (step S105) corresponding to the instruction content different from the instruction content determined in steps S102 to S104 described above is executed. Exit the detection program.
[0030]
When the editing mode is set, the CPU 11 executes the editing program in the ROM 15. As shown in FIG. 6, this editing program is composed of three types of editing subprograms respectively corresponding to velocity expander editing processing, pronunciation duplication editing processing, and timing tuner editing processing.
[0031]
Each of the editing subprograms includes an analyzer and a changer as shown in FIG. Among these, the changer is a program for performing editing processing of performance data, which is the original purpose of each editing subprogram. The analyzer is a program that analyzes performance data to be processed and extracts information necessary for execution of the changer from the performance data, and is executed prior to execution of the changer.
[0032]
The specific processing contents of the analyzer and changer differ depending on each editing subprogram, but the analyzer of any editing subprogram basically stores information necessary for performance data analysis and editing processing according to the flow shown in FIG. Extraction is performed. That is, the MIDI channel number CH is changed from “0” to “15” (steps S201, S205, S206), and whether or not each MIDI channel corresponding to each channel number CH is an object of editing processing, that is, For example, if the analyzer is an analyzer for velocity expander editing processing, it is determined whether or not the content of the register EXP (CH) is “1” (step S202). For the channel number CH for which the determination result is “YES”, the performance data corresponding to the MIDI channel of the channel number CH is loaded into the work area in the RAM 14 (step S203), and the captured performance data is analyzed. Then, information required by the subsequent changer is extracted (step S204). When each process corresponding to the analyzer is completed in this way, each process corresponding to the changer is executed according to the information obtained thereby, and the velocity expander editing process, pronunciation duplication editing process, or timing tuner editing process is completed. It is.
[0033]
Hereinafter, specific processing contents will be described in detail in the order of velocity expander editing processing, pronunciation duplication editing processing, and timing tuner editing processing.
[0034]
(2) Velocity expander editing process
FIG. 9 is a graph in which velocity data corresponding to one MIDI channel (channel number CH) among the MIDI channels designated as processing targets is arranged in the order of performance, and the solid curve P is edited by velocity expander. A change in velocity data before the processing is performed, and a broken curve Q represents a change in velocity data after the velocity expander editing processing. As illustrated in this figure, in the velocity expander editing process, the entire velocity data is within a range limited by the upper limit value set in the register MAX (CH) and the lower limit value set in the register MIN (CH). The distribution range of each velocity data is expanded so as to fit.
[0035]
In the analyzer for the velocity expander editing process, the maximum value max and the minimum value of the velocity data for each MIDI channel designated as a processing target (that is, a MIDI channel with EXP (CH) = “1”). Analysis processing for obtaining min (see FIG. 9) is performed.
[0036]
The direct object of the analysis processing is velocity data, but this velocity data is included in the key-on event that forms part of the performance data. The data format of this key-on event is shown in FIG. As shown in the figure, the key-on event is composed of 3-byte information. First, the first byte indicates the key-on status, and the upper 4 bits represent the identification code “9” of the key-on event, and the lower 4 bits represent the MIDI channel number CH of the key-on event. Next, the second byte represents a note number of a musical sound that is generated by the key-on event. The third byte of the key-on event is velocity data that is a direct processing target of this analysis processing. Therefore, in obtaining the maximum value max and the minimum value min of the velocity data, only the performance data corresponding to the key-on event is focused, and the procedure for determining the magnitude of the velocity data of the third byte is performed. FIG. 11 shows the flow of analysis processing executed according to this procedure. This analysis process corresponds to step S204 shown in FIG. Hereinafter, the analysis process will be described in detail with reference to the flowchart shown in FIG.
[0037]
First, initial setting of the maximum value max and the minimum value min is performed (step S301). Next, the first event is fetched from the performance data corresponding to the MIDI channel to be processed (step S302). Next, it is determined whether or not the value of the upper 4 bits of the first byte of the captured event is “9”, that is, whether or not the event is a key-on event (step S303). If this determination is “YES”, the flow proceeds to step S304, and if “NO”, the flow proceeds to step S308.
[0038]
Next, in step S304, the velocity data Vel at the third byte of the key-on event fetched this time is referred to, and it is determined whether or not the velocity data Vel is equal to or greater than the current maximum value max. If the determination result is “YES”, the maximum value max is updated with the velocity data Vel (step S305), and the process proceeds to step S308. On the other hand, if the determination result in step S304 is “NO”, it is determined whether or not the velocity data Vel of the key-on event captured this time is equal to or less than the current minimum value min (step S306). If the determination result is “YES”, the minimum value min is updated by the velocity data Vel (step S307), and then the process proceeds to step S308. If “NO”, step S307 is not executed and step S308 is not executed. Proceed to
[0039]
Next, in step S308, it is determined whether or not the processing described above has been executed for all events corresponding to the MIDI channel to be processed, that is, whether or not the analysis can be terminated. If the determination result is “NO”, the next event after the currently captured event is captured (step S309), and the processing including steps S303 to S308 is executed for the captured event. In the same manner, the processing including steps S303 to S308 is executed for all events corresponding to the MIDI channel to be processed. When the determination result in step S308 is “YES”, the maximum value max and the minimum value min at that time are stored in a predetermined area in the RAM 14 (step S310), and the process corresponding to the MIDI channel is ended.
[0040]
The changer process is executed according to the flow shown in FIG. First, the channel number CH of the MIDI channel is initialized to “0” (step S401). Next, it is determined whether or not the content of the register EXP (CH) is “1”, that is, whether or not the MIDI channel corresponding to the current channel number CH is the processing target of the velocity expander editing process (step S402). . If the determination result is “NO”, after incrementing the channel number CH by 1 (step S403), it is determined whether the channel number CH exceeds “15” (step S407). If not, the determination in step S402 is performed again.
[0041]
On the other hand, if the content of the register EXP (CH) is “1”, the velocity data Vel of the MIDI channel corresponding to the channel number CH is sequentially read out one by one from the RAM 14 (step S404). Then, the velocity data Vel is updated according to the following equation (1), and the updated velocity data Vel is returned to the original position in the RAM 14 (step S405).

[0042]
However, in this equation (1), max and min are the maximum and minimum values of velocity data in the MIDI channel corresponding to the channel number CH, and are obtained by the processing already described with reference to FIG. This is the information stored in.
[0043]
Next, it is determined whether or not all velocity data corresponding to the MIDI channel of channel number CH has been read (step S406). If the determination result is “NO”, the process returns to step S404, and the same processing is performed on the subsequent velocity data. If the determination result in step S406 is “YES”, after increasing the channel number CH (step S403), the process proceeds to step S407 to determine whether the channel number CH exceeds “15”. to decide. If the determination result is “NO”, the process proceeds to step S402. If another MIDI channel to be processed is found in step S402, the above-described processing is performed for the MIDI channel. When the processes for all the MIDI channels are completed, the determination result in step S407 is “YES”, and the process as the changer is completed.
[0044]
With the above processing, the distribution range of all velocity data corresponding to the processed MIDI channel is expanded within the range limited by the upper and lower limits set in the registers MAX (CH) and MIN (CH). Is done. FIG. 13 illustrates the manner in which the velocity data distribution range is expanded. As shown in this figure, each velocity data distributed in the range from the maximum value max to the minimum value min is the upper limit value while maintaining the mutual magnitude relationship as it is by applying the above formula (1). The distribution range is expanded within a wide range between MAX (CH) and the lower limit value MIN (CH).
[0045]
As described above, according to the present embodiment, the distribution range of velocity data for the specified MIDI channel can be arbitrarily expanded, so that the undulation of the change in volume of a specific part during automatic performance can be arbitrarily increased, Can be given.
[0046]
(3) Overlapping pronunciation editing processing
In the pronunciation duplication editing process, for each MIDI channel to be processed (MIDI channel in which DBL (CH) = “1”), an existing key-on event corresponding to the MIDI channel is Performs processing to add duplicate key-on events. If the duplicate key-on event is added unconditionally, there is a possibility that the sound generation channel will be insufficient. Therefore, after performing analysis processing that grasps the occurrence of existing key-on events in time series, processing is performed to add duplicate key-on events as long as the condition that there is no shortage of pronunciation channels at each time point is satisfied. . The former analysis process is performed by executing the analyzer, and the latter process is performed by executing the changer.
[0047]
In the analysis process of the analyzer (the process corresponding to step S204 in FIG. 8), whether or not there is a shortage of sound generation channels when the same key-on event is added to each key-on event corresponding to the MIDI channel to be processed. Determine. The outline of this process will be described with reference to FIG. 14A and 14B, KON1, KON2,... Each represent a key-on event, and KOFF1, KOFF2,... Represent a key-off event that terminates the sounding operation started by these key-on events. Yes. Each of these key-on events and each key-off event is extracted from the performance data stored in the RAM 14, and is arranged in time series according to each reproduction order, with the horizontal direction of the drawing as the direction of time progression. ing.
[0048]
Here, an example of how to determine the presence / absence of a sound channel shortage for each situation shown in FIGS. 14 (a) and 14 (b) will be described. In the following, it is assumed that the sound source board has four sound generation channels. Also, it is assumed that the illustrated key-on event and key-off event correspond to the MIDI channel that is the processing target of the pronunciation duplication editing process.
[0049]
First, in the situation shown in FIG. 14A, the sound generation periods started by the key-on events KON1 to KON3 do not overlap on the time axis. Accordingly, each of the key-on events KON1 to KON3 can add a duplicate key-on event to each.
[0050]
Next, consider the situation shown in FIG. In this case, some sound generation periods started by the key-on events KON1 to KON5 overlap each other on the time axis. Therefore, it is examined whether or not a duplicate key-on event can be added for each key-on event.
[0051]
First, considering the key-on event KON1, no sound is generated by another key-on event at the timing of occurrence of this key-on event. In addition, key-on events KON2 and KON3 occur before the sounding by the key-on event KON1 is ended by the key-off event KOFF1, but even if sounding corresponding to these key-on events is performed, one sounding channel still remains. Therefore, even if sound generation is performed using one sound generation channel during the sound generation by the key-on event KON1, there is no shortage of sound generation channels. Therefore, for the key-on event KON1, the same key-on event is added, and a determination result is obtained that duplicate sound generation using two sound generation channels for the same sound may be performed.
[0052]
Next, consider the key-on event KON2. According to the result of the above determination, two sound generation channels are used for the key-on event KON1, and duplicate sound generation is performed. Therefore, when sound generation corresponding to the key-on event KON2 is performed, three sound generation channels are used. Become. Then, sound generation corresponding to the key-on event KON3 is started until sound generation by the key-on event KON2 is ended by the key-off event KOFF2. Here, since the sound generation corresponding to the key-on event KON1 is still performed at the sound generation start time corresponding to the key-on event KON3, four sound generation channels are used at this time. Therefore, if sound generation using two sound generation channels is performed for the key-on event KON2, sound generation channels are insufficient when sound generation corresponding to the key-on event KON3 is started. Therefore, for the key-on event KON2, the determination result indicates that duplicate sound generation using two sound generation channels cannot be permitted.
[0053]
  Next, the key-on event KON3 will be examined. If the results of the above determinations are followed, the key-on event KON1 is generated using two sound generation channels, and the key-on event KON2 is sounded using one sound channel. When sounding corresponding to is performed, four sounding channels are used. Therefore, for the key-on event KON3, there are two sound channels.useObviously, you can't make the same pronunciation.
[0054]
Next, consider the key-on event KON4. At the point of time when the sound corresponding to the key-on event KON4 is started, the sound corresponding to the key-on event KON1 has already ended. Therefore, at this time, three sound generation channels are used to generate sound corresponding to the key-on events KON2, KON3, and KON4, and one sound generation channel remains. Further, until the sound generation by the key-on event KON4 is ended by the key-off event KOFF4, the sound generation corresponding to another key-on event is not started. Therefore, for the key-on event KON4, a determination result is obtained that it is possible to recognize duplicate pronunciation using two tone generation channels.
[0055]
In the above-described determination, if sound generation periods corresponding to a plurality of key-on events overlap on the time axis, and there is a shortage of sound channels when sound generation is overlapped for any of the key-on events, playback is performed first in time. This is performed in accordance with a rule that allows duplicate pronunciation with priority given to key-on events. In addition to such rules, various rules for avoiding conflicts between key-on events for overlapping pronunciations can be considered, but the description thereof is omitted.
[0056]
In the analysis process of the pronunciation duplication editing process, according to the principle as described above, whether to apply the duplicate pronunciation is determined for each key-on event corresponding to the MIDI channel to be processed.
[0057]
In order to perform this analysis process, it is necessary to grasp the occurrence state of each key-on event and each key-off event in time series. Therefore, in the present embodiment, the step time included in the performance data is referred to determine the occupied position on the time axis of the sound generation period corresponding to each key-on event. That is, data representing the step time is added before each event in the performance data as illustrated in FIG. This step time is a waiting time from the reproduction of the event immediately before it to the reproduction of the event (the event immediately after the step time). Therefore, by calculating the total value of all the step times existing before an event for an arbitrary event, it is possible to obtain the relative reproduction time of the event based on the start time of the automatic performance. In this embodiment, by performing such calculation for each key-on event and each key-off event, the occupied position on the time axis of the sound generation period corresponding to each key-on event is obtained.
[0058]
Next, a specific flow of the analysis process of the pronunciation duplication editing process will be described with reference to FIG. First, the first event is fetched from the performance data corresponding to the MIDI channel to be processed (step S501). Next, it is determined whether or not the captured event is a key-on event (step S502). If this determination is “YES”, the flow proceeds to step S503, and if “NO”, the flow proceeds to step S508.
[0059]
Next, when proceeding to step S503, the peripheral events of the key-on event captured this time are investigated. That is, as described above, paying attention to each sound generation period corresponding to each key-on event grasped based on the step time, the following items are investigated.
[0060]
a. Whether the pronunciation period corresponding to the key-on event imported this time does not overlap on the time axis with the pronunciation period corresponding to other key-on events.
b. Whether or not there is a shortage of pronunciation channels when overlapping pronunciations are applied to the key-on event imported this time when overlapping pronunciation periods are recognized.
[0061]
Next, the process proceeds to step S504, and it is determined whether or not duplicate pronunciation can be applied to the key-on event captured this time based on the investigation result in step S503. The result of this determination can be divided into the following three types.
[0062]
a. It does not matter to allow duplicate pronunciation unconditionally.
The situation in which this determination is made is illustrated in FIG. In this example, the sound generation periods corresponding to the key-on events KON1 and KON2 do not overlap at all with the sound generation periods corresponding to other key-on events. Therefore, when the key-on event captured this time is placed in a situation such as the key-on events KON1 and KON2 shown in this example, a determination result is obtained that duplicate pronunciation can be applied unconditionally.
[0063]
b. Conditional duplicate pronunciation may be allowed. That is, since there are other key-on events that compete with the key-on event captured this time, whether or not the duplicate pronunciation can be applied to the key-on event captured this time depends on whether or not the duplicate pronunciation is applied to these other key-on events.
[0064]
The situation in which this determination is made is illustrated in FIG. In this example, the key-on events KON1, KON2, and KON3 have their respective sound generation periods overlapping on the time axis, and therefore it is allowed to apply overlap sounding to only one of them. Further, the key-on event KON4 has the same sound generation period as the key-on events KON2 and KON3. Accordingly, for the key-on event KON4, duplicate pronunciation may be recognized only when duplicate pronunciation is not applied to any of the key-on events KON2 and KON3. If the key-on event captured this time is placed in a situation such as the key-on events KON1 to KON4 shown in this example, whether or not the duplicate pronunciation can be applied to these is determined by the duplicate pronunciation for other competing key-on events. It depends on the application.
[0065]
c. There is no room to allow duplicate pronunciation in the key-on event that was imported this time.
Even if the key-on event is not added, all the sound generation channels are already used, and there is always a shortage of sound generation channels when duplicate sound generation is performed.
[0066]
If it is determined in step S504 that the determination result corresponds to a or b, the process proceeds to step S505. If the determination result corresponds to c, the process proceeds to step S508.
[0067]
Next, in step S505, it is determined whether or not the determination result in step S504 is b. If this determination is “YES”, the flow proceeds to step S 506, and information indicating the content of the key-on event captured this time and the position on the time axis is recorded in the RAM 14. Then, the process proceeds to step S508.
[0068]
On the other hand, if the determination result in step S504 is “NO”, that is, if the key-on event captured this time is placed in the above-described situation b, the process proceeds to step S507, and adjustment is performed to avoid contention. For example, this adjustment is performed according to a rule for applying overlapping pronunciations with priority given to a key-on event that is reproduced first, as already described with reference to FIG. In addition to this method, for example, competing key-on events may be displayed on the CRT display 2, and the user may be allowed to select which key-on event the overlapping pronunciation is applied to. When the key-on event to which the overlapping pronunciation is applied is determined, information indicating the content of the key-on event and the position on the time axis is recorded in the RAM 14, and the process proceeds to step S508.
[0069]
Next, in step S508, it is determined whether or not the processing described above has been executed for all events corresponding to the MIDI channel to be processed, that is, whether or not the analysis can be terminated. If the determination result is “NO”, the next event is captured (step S509), and the process returns to step S502. In the same manner, the processing including steps S502 to S508 is executed for all events corresponding to the MIDI channel to be processed. Then, when the determination result in step S508 is “YES”, the processing as the analyzer is terminated.
[0070]
The changer process is executed according to the flow shown in FIG. First, the channel number CH of the MIDI channel is initialized to “0” (step S601). Next, it is determined whether or not the content of the doubler register DBL (CH) is “1”, that is, whether or not the MIDI channel corresponding to the current channel number CH is the processing target of the pronunciation duplication editing process (step S602). ). If the determination result is “NO”, after incrementing the channel number CH by 1 (step S605), it is determined whether the channel number CH exceeds “15” (step S604). If not, the determination in step S602 is performed again.
[0071]
On the other hand, if the content of the register DBL (CH) is “1”, the process proceeds to step S603 to execute a key-on event insertion process. That is, the MIDI channel events corresponding to the channel number CH are sequentially read from the RAM 14 one by one, and the read events are sequentially written to another area in the RAM 14. At that time, the information on each key-on event stored in the RAM 14 when the analyzer is executed is referred to, and it is determined whether or not each event read from the RAM 14 is a key-on event to which duplicate pronunciation is applied. . Then, the key-on event having the same contents as the key-on event to which the duplicate pronunciation is applied is added to the RAM 14 and written. When this key-on event insertion process ends, the process proceeds to step S605, and the channel number CH is incremented by one. Then, the process proceeds to step S604, and it is determined whether or not the channel number CH has exceeded “15”. If this determination is “NO”, the flow returns to step S602. Thereafter, the process proceeds and the channel number CH exceeds “15”, so that the determination result in step S604 is “YES”, and the process as the changer ends.
[0072]
As described above, according to the present embodiment, it is possible to apply overlapping pronunciation to a key-on event of a desired MIDI channel, so that the sound of a desired part can be made thick with a core.
[0073]
(4) Timing tuner editing process
First, the outline of the timing tuner editing process will be described with reference to FIG. In FIG. 19, the horizontal straight line indicates the time axis, and a series of white triangle marks arranged above the time axis indicate the reproduction timing of the key-on event to be processed by this timing tuner editing process. Show. In addition, a series of black triangle marks are written on the lower side of the time axis, and these indicate timings when the user wants to pronounce.
[0074]
In this timing tuner editing process, each key-on event or key-off event corresponding to the MIDI channel to be processed (MIDI channel with TUN (CH) = “1”) is reproduced at the timing desired by the user. Thus, the reproduction timing is corrected. The timing tuner editing process includes an analysis process for obtaining a time difference between the playback timing of the existing key-on event or key-off event and the playback timing desired by the user, that is, Δt1, Δt2,... Shown in FIG. And a process for correcting the reproduction timing of each key-on event and key-off event. The program for performing the former analysis processing is an analyzer, and the program for performing the latter processing is a changer.
[0075]
In executing the analyzer, it is necessary to prepare in advance data representing the reproduction timing desired by the user. There are the following two methods for obtaining this data.
[0076]
a. Method by the user performing a real performance
For example, a user plays a specific part (for example, a rhythm part) using another electronic musical instrument. The actual performance data obtained at this time is supplied from the MIDI interface 17 to the performance data editing apparatus, and the reproduction timing of the key-on event or key-off event in the actual performance data is used as the reproduction timing desired by the user.
[0077]
b. As shown in FIG. 20, a method for creating actual performance data for designating a desired reproduction timing by hitting the space key of the keyboard 3 by the user.
When this method is used, the user performs a rhythm performance using only the space key after setting the operation mode of the performance data editing apparatus to the MIDI data creation mode. In other words, the space key may be turned on at the timing at which sounding is to be started, and the space key may be turned off at the time at which sounding is to be ended. Therefore, this method is convenient for a user who is not familiar with playing electronic musical instruments.
[0078]
FIG. 21 shows a flow of an actual performance data creation program for creating actual performance data representing the reproduction timing of a key-on event of a rhythmic musical tone according to this method. The processing contents of this program will be described below.
[0079]
This program is started by inputting a predetermined command. First, in step S701, it is determined whether there is a space key on event. If this determination is “NO”, the flow proceeds to step S 704, where it is determined whether an instruction to end the process has been given (whether an end button (not shown) has been pressed). If this determination is “NO”, the flow returns to step S701.
[0080]
When the space key ON event occurs, the determination result in step S701 is “YES”, and the flow advances to step S702. Then, it is determined whether or not the current operation mode is the MIDI data creation mode. If the determination result is “NO”, the program is terminated without creating any data.
[0081]
On the other hand, if the determination result in step S702 is “YES”, the process advances to step S703 to execute the MIDI encoding process. That is, a step time corresponding to the elapsed time from the start of execution of this program and a predetermined key-on event are created and written in a predetermined area in the RAM 14. Then, after this writing is completed, the process returns to step S701.
[0082]
Thereafter, when the space key ON event occurs again, the process proceeds to steps S703 via steps S701 and S702, and the MIDI encoding process is executed again. In this case, a step time and a predetermined key-on event corresponding to the elapsed time from the previous writing of the step time and the key-on event to the present are created and written in a predetermined area in the RAM 14. Thereafter, similarly, every time a space key on event is detected, the step time and the key on event are recorded in the RAM 14. When the operation of the space key is interrupted and an instruction to end the process is given, the determination result in step S704 is “YES”, and the process ends.
[0083]
Through the above processing, actual performance data representing the reproduction timing of the key-on event desired by the user is obtained in the RAM 14.
[0084]
When data for designating the playback timing of both the key-on event and the key-off event of a continuous sound is to be created by the space key operation, not only the space key on event but also the off event for the program shown in FIG. It is sufficient to add processing so as to also determine.
[0085]
When the actual performance data is obtained in this way, the analyzer is executed next, and the MIDI channel (MIDI channel where TUN (CH) = “1”) that is the target of the timing tuner editing process is The analysis process shown in FIG. 22 is executed. First, in step S801, musical score data from which events other than key-on events have been removed from performance data in the RAM 14 is created. That is, a series of events corresponding to the MIDI channel to be processed and the step times inserted between these events are sequentially read from the RAM 14. In parallel with this reading, the following processing is performed.
[0086]
a. Until the key-on event is read, the step time read during that time is accumulated.
b. When the key-on event is read out, the accumulated step time value and the key-on event are written into the storage area for the score data in the RAM 14.
c. After this writing, the accumulated value of the step time is returned to 0 and the process a is performed.
[0087]
As a result of the above processing, at the time when all the events corresponding to the MIDI channel to be processed are read, score data that does not include any event other than the key-on event is obtained in the RAM 14.
[0088]
In step S802, the event and the step time set before it are sequentially extracted from each of the score data and the actual performance data in the RAM. In step S803, each step time of the musical score data extracted in this way and each step time of the actual performance data are compared with each other, and the differences (that is, Δt1, Δt2 shown in FIG. 19) are compared. , ..., etc.). Then, this difference is written in the RAM 14 in correspondence with each key-on event, and the analysis process is terminated.
[0089]
The changer process is executed according to the flow shown in FIG. First, the channel number CH of the MIDI channel is initialized to “0” (step S901). Next, it is determined whether or not the content of the tuner register TUN (CH) is “1”, that is, whether or not the MIDI channel corresponding to the current channel number CH is the processing target of the timing tuner editing process (step S902). . If the determination result is “NO”, the channel number CH is incremented by 1 (step S905), and the process proceeds to step S904 to determine whether the channel number CH exceeds “15”. If this determination is “NO”, the process of step S902 is performed again.
[0090]
On the other hand, when the content of the register TUN (CH) is “1”, the process proceeds to step S903 to execute a step time change process. That is, the MIDI channel events corresponding to the channel number CH are sequentially read from the RAM 14 one by one, and the read events are sequentially written to another area in the RAM 14. During this time, when a key-on event is read out, the difference in step time corresponding to the key-on event stored in the RAM 14 is referred to. The step time immediately before the key-on event is corrected by a value corresponding to the difference, and the corrected step time is written in the RAM 14 together with the key-on event. As a result of this processing, the reproduction timing of each key-on event corresponding to the MIDI channel to be processed is corrected to coincide with the reproduction timing of each key-on event of the actual performance data. The difference is not used as it is for correction, but a difference obtained by multiplying the difference by a predetermined coefficient smaller than 1 is used for correction so that the reproduction timing of the key-on event approaches the key-on timing of the actual performance data. Good.
[0091]
When this step time changing process is completed, the process proceeds to step S905, and the channel number CH is increased. The process advances to step S904 to determine whether or not the channel number CH has exceeded “15”. If this determination is “NO”, the flow returns to step S902. Thereafter, when the process proceeds and the channel number CH exceeds “15”, the determination result in step S904 is “YES”, and the process as the changer is terminated.
[0092]
As described above, according to the present embodiment, it is possible to finely adjust the reproduction timing of the musical sound corresponding to a specific part to the timing desired by the user by the timing tuner editing process. You can have your favorite ride.
[0093]
【The invention's effect】
  As explained above, according to the present invention,TheDuring dynamic performance, the desired part can be played without causing the harmful effect that other musical sounds that should be pronounced are not pronounced due to insufficient number of sound channels.aboutThick sound with a coreEdit performance data so thatcan do.
[Brief description of the drawings]
FIG. 1 is a diagram showing an appearance of a performance data editing apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram showing an electrical configuration of the embodiment.
FIG. 3 is a diagram showing a control register according to the embodiment;
FIG. 4 is a diagram showing a control program in a ROM 15 of the same embodiment.
FIG. 5 is a flowchart showing the operation of the embodiment.
FIG. 6 is a flowchart showing the operation of the embodiment.
FIG. 7 is a flowchart showing the operation of the embodiment.
FIG. 8 is a flowchart showing the operation of the embodiment.
FIG. 9 is a diagram illustrating a velocity expander editing process according to the embodiment.
FIG. 10 is a diagram showing a key-on event that is a processing target of velocity expander editing processing;
FIG. 11 is a flowchart showing the operation of the embodiment.
FIG. 12 is a flowchart showing the operation of the embodiment.
FIG. 13 is a diagram illustrating an effect of a velocity expander editing process.
FIG. 14 is a diagram illustrating pronunciation duplication editing processing according to the embodiment;
FIG. 15 is a diagram illustrating a processing target of pronunciation duplication editing processing;
FIG. 16 is a flowchart showing the operation of the embodiment.
FIG. 17 is a diagram for explaining analysis processing of pronunciation duplication editing processing;
FIG. 18 is a flowchart showing the operation of the embodiment.
FIG. 19 is a diagram illustrating timing tuner editing processing according to the embodiment;
FIG. 20 is a diagram illustrating a method for creating actual performance data in the embodiment.
FIG. 21 is a flowchart showing the operation of the embodiment.
FIG. 22 is a flowchart showing the operation of the embodiment.
FIG. 23 is a flowchart showing the operation of the embodiment.
[Explanation of symbols]
11: CPU (performance data adding means),
12 …… Display section, 13 …… Operation section (part designation means),
14: RAM (performance data storage means),
15 ... ROM, 16 ... hard disk drive,
17 …… MIDI interface,
6: Sound board, 5L, 5R: Speaker.

Claims (2)

A performance data editing device for editing performance data for each part to be generated,
A specifying unit that specifies one part from a plurality of parts in the performance data,
The performance data is analyzed to investigate the tone generation period of each musical tone in the part designated by the designation means , and the maximum of the musical sounds in the part that are simultaneously producing sound during the musical tone production period of one musical tone in the part. And adding means for adding, to the performance data, data for generating a musical tone having the same sounding period as the sounding period of the first musical sound when the maximum number is less than a predetermined number. A performance data editing apparatus characterized by that. The adding means adds, to the performance data, data for generating the same musical tone as the first musical tone in the same pronunciation period as the first musical tone when the maximum number is less than a predetermined number.
The performance data editing apparatus according to claim 1, wherein:

Images