JPH0642158B2 - Speech synthesizer - Google Patents
Speech synthesizerInfo
- Publication number
- JPH0642158B2 JPH0642158B2 JP58205227A JP20522783A JPH0642158B2 JP H0642158 B2 JPH0642158 B2 JP H0642158B2 JP 58205227 A JP58205227 A JP 58205227A JP 20522783 A JP20522783 A JP 20522783A JP H0642158 B2 JPH0642158 B2 JP H0642158B2
- Authority
- JP
- Japan
- Prior art keywords
- articulatory
- waveform
- voice
- phoneme
- string
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
- 230000002194 synthesizing effect Effects 0.000 claims description 12
- 238000000034 method Methods 0.000 description 9
- 210000000056 organ Anatomy 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000001308 synthesis method Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 210000003254 palate Anatomy 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 210000002105 tongue Anatomy 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 210000001260 vocal cord Anatomy 0.000 description 2
- 239000002131 composite material Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 210000000088 lip Anatomy 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Electrophonic Musical Instruments (AREA)
Description
【発明の詳細な説明】 本発明は音声合成装置に関する。The present invention relates to a speech synthesizer.
従来、単音節の半分程度の時間長を有するＣＶあるいは
ＶＣ（ここでＣは子音、Ｖは母音を表わす）に対応した
自然音声波形をあらかじめ用意しておき、入力文字列に
従って前記ＣＶあるいはＶＣに対応する波形を編集合成
することにより音声波形を生成する方式が、下記資料
(1)等により知られている。Conventionally, a natural speech waveform corresponding to a CV or VC (where C is a consonant and V is a vowel) having a time length about half that of a single syllable is prepared in advance, and the CV or VC is assigned to the CV or VC according to the input character string. The method of generating a voice waveform by editing and synthesizing the corresponding waveforms
(1) Known by others.
三留、伏木田、「ＣＶ，ＶＣ波形のピッチ同期的補間に
よる任意語合成方式」 日本音響学会、音声研究会資料、Ｓ８２−０６（１９８
２−４）しかしながら、前記の方式においては高品質な
合成音声が得られるがＣＶ，ＶＣ等に対応する音声波形
の情報量が大きく膨大な容量の記憶装置を必要とする欠
点があった。Mitome, Fushikida, "Arbitrary Word Synthesis Method by Pitch-Synchronous Interpolation of CV and VC Waveforms", Acoustical Society of Japan, Speech Study Group, S82-06 (198)
2-4) However, the above-mentioned method has a drawback that a high-quality synthesized speech can be obtained, but a storage device having a large amount of information of a speech waveform corresponding to CV, VC, etc. is required and a huge capacity is required.
本発明の目的は、あらかじめ用意すべき音声波形等の音
声データのメモリ容量が比較的少なくて済む音声合成装
置を提供することにある。It is an object of the present invention to provide a voice synthesizer that requires a relatively small memory capacity for voice data such as voice waveforms to be prepared in advance.
本発明によれば、入力文字列を音声に変換する型の音声
合成装置において、 前記入力文字列を音韻あるいは音素の時間長よりも短い
時間区間に対応する調音記号で構成され該調音記号が異
なる音素系列に従って発声された音声波形の中から調音
様式の類似する部分の音声波形で代用できる調音記号列
に変換する手段と、 前記調音記号に対応する音声波形を記憶する記憶回路
と、 前記変換された調音記号列に従って対応する音声波形を
前記記憶回路から引き出し編集合成することにより、前
記入力文字列に対応する音声波形を生成する手段とを有
することを特徴とする音声合成装置が得られる。According to the present invention, in a speech synthesizer of a type that converts an input character string into speech, the input character string is composed of articulatory symbols corresponding to a time section shorter than a time length of a phoneme or a phoneme, and the articulatory symbols are different. Means for converting into an articulatory symbol string that can be substituted by a voice waveform of a part having a similar articulation style from a voice waveform uttered according to a phoneme sequence; a storage circuit that stores a voice waveform corresponding to the articulatory symbol; And a means for generating a voice waveform corresponding to the input character string by extracting and editing and synthesizing a corresponding voice waveform from the storage circuit according to the articulatory symbol string.
本発明の特徴は調音記号に対応した音声波形等を単位音
声データとしてあらかじめ用意しておき、入力文字列か
ら変換された調音記号列に従って前記単位音声データを
編集合成することにより、前記あらかじめ用意すべき単
位音声データのメモリ容量が比較的少なくと済むことに
ある。The feature of the present invention is that the voice waveform corresponding to the articulatory symbols is prepared in advance as the unit voice data, and the unit voice data is edited and synthesized according to the articulatory symbol string converted from the input character string to prepare in advance. This is because the memory capacity of the unit voice data to be used is relatively small.
人間の調音器官は第１図(a)に示されるように、声帯、
舌、唇、口蓋帆等があり、これらの調音器官が神経パル
ス信号に従って制御されることにより種々の音声が生成
される。従って調音器官の動きが類似していれば、類似
した音声波形が生成される。さらに、これらの調音器官
の動きを表わす調音パラメータ値が近ければ生成される
音声波形も類似していることは明らかである。さて、従
来の前記ＣＶ，ＶＣ波形編集型の合成方式ではＣＶ，Ｖ
Ｃ名に対応して音声波形を用意していたが、調音パラメ
ータの動きということから考えると、かなり冗長な波形
が含まれている。例えば、前記ＣＶ，ＶＣ波形編集方式
では／ｋａ／に対する音声波形と／ｇａ／に対する音声
波形はそれぞれ別々に用意されていた。しかしながら／
ｋａ／に対する調音器官の動きと／ｇａ／に対する調音
器官の動きは非常に似ており、子音／ｋ／、と／ｇ／に
おいて舌と口蓋等によってつくられるせばめの位置はほ
とんど同じであり、主な相違は子音部において声帯が振
動しているか否か（有声か無声か）ということにある。
従って、／ｋａ／における子音部／ｋ／の無声区間の後
の有声区間（母音／ａ／の定常部に移行する区間で第１
図(b)の１０３に対応する）は調音パラメータ値はほぼ
同一であり、／ｇａ／の対応する部分波形（第１図(b)
の１０８に対応）を用いて近似できる。一般に、／ｋＶ
／−／ｇＶ／，／ｔＶ／−／ｄＶ／，／ｐＶ／−／ｂＶ
／（ここでＶは母音を表わす）の対についても同様に母
音へ移行する部分の波形が互いに共用できることは明ら
かである。Human articulatory organs, as shown in FIG. 1 (a), are vocal cords,
There are tongues, lips, palate, etc., and various sounds are generated by controlling these articulatory organs according to nerve pulse signals. Therefore, if the movements of the articulatory organs are similar, similar speech waveforms are generated. Further, it is obvious that the generated voice waveforms are similar if the articulatory parameter values representing the movements of these articulatory organs are close to each other. Now, in the conventional CV, VC waveform editing type synthesis method, CV, V
Although a voice waveform was prepared corresponding to the C name, a considerably redundant waveform is included in view of the movement of the articulatory parameters. For example, in the CV and VC waveform editing methods, the voice waveform for / ka / and the voice waveform for / ga / are prepared separately. However/
The movement of the articulatory organ for ka / and the movement of the articulatory organ for / ga / are very similar, and the positions of the fits created by the tongue and the palate are almost the same in the consonants / k /, and / g /. The major difference is whether or not the vocal cords are vibrating in the consonant part (voiced or unvoiced).
Therefore, in the voiced section after the unvoiced section of the consonant part / k / in / ka / (the section that shifts to the stationary part of the vowel / a /
(Corresponding to 103 in FIG. 2B), the articulatory parameter values are almost the same, and the corresponding partial waveform of / ga / (FIG. 1B)
(Corresponding to No. 108)). Generally, / kV
/-/ GV /, / tV /-/ dV /, / pV /-/ bV
It is clear that the same can be said for the pair of / (where V is a vowel), and the waveforms of the transitions to vowels can be shared with each other.
ここでは、第１図(b)に例示される如く、調音様式を考
慮して定められる。ＣＶ，ＶＣ波形よりも短かく、異な
る音素系列に従って発声された音声波形により代用され
得る時間区間を調音セグメントと呼び、調音セグメント
において用いられる音声波形を調音素片波形と呼ぶ。Here, as illustrated in FIG. 1 (b), it is determined in consideration of the articulation style. A time section that is shorter than the CV and VC waveforms and that can be substituted by a voice waveform uttered according to a different phoneme sequence is called an articulation segment, and a voice waveform used in the articulation segment is called an articulator segment waveform.
以上の例からも明らかなように、調音様式が同じである
調音セグメントに対しては同じ調音記号を用いて表わ
し、この調音記号に対応した調音素片波形をあらかじめ
用意しておき、入力として与えられる文字列を調音記号
列に変換した後、この調音記号列に従って前記調音記号
に対応した調音素片波形を編集合成すれば比較的少ない
音声波形メモリで任意音声の合成（規則合成）が可能で
ある。入力文字列を調音記号列に変換するためには、例
えば、まず、入力文字列を音素列に変換した後、前記音
素列を音素列に対してあらかじめ用意される調音記号列
テーブル（辞書）を用いて調音記号列に変換することが
できる。As is clear from the above example, articulatory segments with the same articulation style are represented using the same articulatory symbol, and the articulatory element waveform corresponding to this articulatory symbol is prepared in advance and given as an input. After converting the character string to be converted into an articulatory symbol string, by editing and synthesizing the articulatory segment waveforms corresponding to the articulatory symbol according to this articulatory symbol string, it is possible to synthesize arbitrary voices (rule synthesis) with a relatively small voice waveform memory. is there. In order to convert the input character string into the articulatory symbol string, for example, first, the input character string is converted into a phoneme string, and then the phoneme string is prepared in advance with respect to the phoneme string by an articulatory symbol string table (dictionary). Can be used to convert into a string of articulatory symbols.
例えば、「貸車」という文字列が入力された場合には、
まず／ｋａｓｙａ／という音素列に変換した後、さらに
／＊−ｋ(g)ａ−ａ(z)−ｓ(i)−(h)ｙａ／という調音記
号列に変換する。ここで、＊は声道閉鎖区間（無音部）
を表わし、ｋは無声破裂部を表わし、(g)ａとａ(z)はそ
れぞれ／ｇａ／と／ａｚ／の過渡部の一部を表わし、ｓ
(i)は／ｓｉ／の無声まさつ部を表わし、(h)ｙａは／ｈ
ｙａ／の有声過渡部を表わし調音記号である。For example, if the character string "rental car" is entered,
First, it is converted into a phoneme string of / kasya /, and then converted into an articulatory symbol string of / *-k (g) aa (z) -s (i)-(h) ya /. Where * is the vocal tract closed section (silent part)
, K represents a silent burst part, (g) a and a (z) represent a part of the transient part of / ga / and / az /, respectively, and s
(i) represents the silent portion of / si /, and (h) ya is / h
It is an articulatory symbol that represents the voiced transient portion of ya /.
さらに、本例を第１図（ｂ）、（ｃ）、（ｄ）を用いて
詳細に説明する。第１図（ｂ）、（ｃ）は本例に用いる
音声波形を含む音声波形例を表す。また、（ｄ）は本願
の方式により本例（「貸車」）を編集合成した際の調音
素片波形の接続順序を示す。（ｄ）に示されるように
「貸車」という音声波形は、（ｂ）、（ｃ）における調
音素片波形を１０１、１０２、１０８、１１０、１１
２、１１７の順序で接続することにより得られる。Further, this example will be described in detail with reference to FIGS. 1 (b), (c), and (d). FIGS. 1B and 1C show examples of voice waveforms including the voice waveform used in this example. Further, (d) shows a connection order of articulatory element waveforms when this example (“rental car”) is edited and synthesized by the method of the present application. As shown in (d), the voice waveform “rental car” is the articulatory unit waveforms 101, 102, 108, 110, 11 in (b) and (c).
It is obtained by connecting in the order of 2, 117.
本例からも明らかなように、従来の前記ＣＶ、ＶＣ波形
編集型の合成方式によれば、第１図（ｂ）、（ｃ）の波
形を全て用意しておく必要があるのに対して、本方式に
よれば、１０３、１１４、１１５等の波形は他の調音素
片波形により代用できるため、あらかじめ用意する必要
がなく、比較的小容量の記憶回路により音声合成装置が
実現できる。As is clear from this example, according to the conventional CV and VC waveform editing type synthesis method, it is necessary to prepare all the waveforms of FIGS. 1B and 1C. According to this method, the waveforms of 103, 114, 115, etc. can be substituted by other articulatory element waveforms, so that it is not necessary to prepare in advance, and a speech synthesizer can be realized by a storage circuit of a relatively small capacity.
なお、以上の説明において、Ｃ（Ｖ）、（Ｃ）Ｖのよう
に（ ）を用いて表される調音素片記号はＣＶという音
声波形から（ ）で囲まれた音素部分を除いた波形に対
応する。In the above description, an articulatory segment symbol represented by using () such as C (V) and (C) V is a waveform obtained by removing the phoneme portion surrounded by () from the voice waveform of CV. Correspond.
また、あらかじめ用意すべき調音素片波形は、例えば、
人間の発声した音声波形から音声学的知識を用いて対応
する区間の波形を切り出しすることにより得ることがで
きる。Further, the articulatory element waveform to be prepared in advance is, for example,
It can be obtained by cutting out the waveform of the corresponding section from the speech waveform uttered by a human using the phonetic knowledge.
前述の如く、本発明においては、ＣＶ、ＶＣ波形編集合
成方式におけるＣＶ、ＶＣ波形等の単位音声波形よりも
時間的に短かい単位音声を用いており、波形のメモリ容
量が少なくて済むばかりでなく、調音器官の動きも正確
に反映しているため高品質な合成音声を得ることが可能
である。As described above, in the present invention, the unit voices which are shorter in time than the unit voice waveforms such as the CV and VC waveforms in the CV and VC waveform editing / synthesizing method are used, and the memory capacity of the waveforms is small. In addition, since the movement of the articulatory organ is accurately reflected, it is possible to obtain high-quality synthetic speech.
次に図面を用いて本発明を詳細に説明する。Next, the present invention will be described in detail with reference to the drawings.
第２図は本発明の実施例を示すブロック図である。ま
ず、文字列が文字列入力端子２０１を介して文字音素記
号変換回路２０２に入力される。文字音素記号変換回路
２０２は前記文字列を音素列に変換し、音素列データ伝
送路２０３を介して音素列調音記号列変換回路２０４に
出力する。音素列調音記号列変換回路２０４は前記音素
列を調音セグメント毎に与えられる調音記号の系列に変
換し、調音記号列伝送路２０５を介して波形編集合成回
路２０６に出力する。波形編集合成回路２０６は前記調
音記号列に従って該調音記号に対応する調音素片波形の
アドレスデータを生成し、調音素片波形記憶回路２０７
に出力する。調音素片波形記憶回路２０７は前記アドレ
スデータに従って、対応する調音素片波形を波形データ
伝送路２０９を介して波形編集合成回路２０６に出力す
る。波形編集合成回路２０６は前記調音素片波形を編集
合成し合成波形を合成波形出力端子２１０を介して出力
する。前記音声波形を編集合成する方式としては、例え
ば、前記調音素片波形の接続部分をピッチ同期的に補間
して編集合成すれば良いが、この方式は前記資料(1)に
詳しいのでここでは説明を省略する。FIG. 2 is a block diagram showing an embodiment of the present invention. First, a character string is input to the character phoneme symbol conversion circuit 202 via the character string input terminal 201. The character-phoneme symbol conversion circuit 202 converts the character string into a phoneme string and outputs it to the phoneme string articulatory symbol string conversion circuit 204 via the phoneme string data transmission path 203. The phoneme string articulatory symbol string conversion circuit 204 converts the phoneme string into a series of articulatory symbols given to each articulatory segment, and outputs it to the waveform editing / synthesizing circuit 206 via the articulatory symbol string transmission path 205. The waveform editing / synthesizing circuit 206 generates address data of an articulatory element waveform corresponding to the articulatory symbol according to the articulatory symbol string, and the articulatory element waveform storage circuit 207.
Output to. The articulation segment waveform storage circuit 207 outputs the corresponding articulation segment waveform to the waveform editing / synthesizing circuit 206 via the waveform data transmission path 209 in accordance with the address data. The waveform editing / synthesizing circuit 206 edits and synthesizes the articulatory segment waveforms and outputs the synthesized waveform via the synthesized waveform output terminal 210. As a method of editing and synthesizing the voice waveform, for example, the connected portion of the articulatory segment waveform may be interpolated in a pitch-synchronous manner to edit and synthesize, but since this method is described in detail in the material (1), it will be described here. Is omitted.
以上の説明においては調音記号に対応する調音素片波形
を編集合成するものとしたが、調音記号に対応するホル
マントパラメータ等の言わゆる特徴パラメータを用いて
音声合成を行なう方式に対しても、同様のメモリ節減効
果が得られることは明らかである。In the above description, the articulatory segment waveforms corresponding to the articulatory symbols are edited and synthesized, but the same applies to the method of performing voice synthesis using so-called characteristic parameters such as formant parameters corresponding to the articulatory symbols. It is clear that the memory saving effect of is obtained.
第１図(a)，(b)は調音パラメータを説明するための説明
図で同図(a)は調音器官の概念図、同図(b)は調音セグメ
ントの例を示す図 第２図は本発明の実施例を示すブロック図である。図に
おいて、 ２０１は文字列入力端子、２０２は文字音素記号変換回
路、２０３は音素列データ伝送路、２０４は音素列調音
記号列変換回路、２０５は調音記号列伝送路、２０６は
波形編集合成回路、２０７は調音素片波形記憶回路、２
０８はアドレスデータ伝送路、２０９は波形データ伝送
路、２１０は合成波形出力端子である。FIGS. 1 (a) and 1 (b) are explanatory diagrams for explaining articulatory parameters. FIG. 1 (a) is a conceptual diagram of an articulatory organ, and FIG. 1 (b) is an example of an articulatory segment. It is a block diagram which shows the Example of this invention. In the figure, 201 is a character string input terminal, 202 is a character phoneme symbol conversion circuit, 203 is a phoneme string data transmission path, 204 is a phoneme string articulation symbol string conversion circuit, 205 is an articulation symbol string transmission path, and 206 is a waveform editing / synthesizing circuit. , 207 are articulatory element waveform storage circuits, 2
Reference numeral 08 is an address data transmission line, 209 is a waveform data transmission line, and 210 is a composite waveform output terminal.
Claims (1)
装置において、 前記入力文字列を音韻あるいは音素の時間長よりも短い
時間区間に対応する調音記号で構成され該調音記号が異
なる音素系列に従って発声された音声波形の中から調音
様式の類似する部分の音声波形で代用できる調音記号列
に変換する手段と、 前記調音記号に対応する音声波形を記憶する記憶回路
と、 前記変換された調音記号列に従って対応する音声波形を
前記記憶回路から引き出し編集合成することにより、前
記入力文字列に対応する音声波形を生成する手段とを有
することを特徴とする音声合成装置。1. A speech synthesizer for converting an input character string into speech, wherein the input character string is composed of articulatory symbols corresponding to a time section shorter than the time length of a phoneme or a phoneme, and the phoneme symbols having different articulatory symbols. Means for converting from a voice waveform uttered according to a sequence into a sequence of articulatory symbols that can be substituted by a voice waveform of a portion having a similar articulation style; a storage circuit for storing a voice waveform corresponding to the articulatory symbol; A voice synthesizing device, comprising means for generating a voice waveform corresponding to the input character string by extracting a corresponding voice waveform from the storage circuit according to an articulatory symbol string, and editing and synthesizing the voice waveform.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58205227A JPH0642158B2 (en) | 1983-11-01 | 1983-11-01 | Speech synthesizer |
| EP84113186A EP0144731B1 (en) | 1983-11-01 | 1984-11-02 | Speech synthesizer |
| DE8484113186T DE3473956D1 (en) | 1983-11-01 | 1984-11-02 | Speech synthesizer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58205227A JPH0642158B2 (en) | 1983-11-01 | 1983-11-01 | Speech synthesizer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS6097396A JPS6097396A (en) | 1985-05-31 |
| JPH0642158B2 true JPH0642158B2 (en) | 1994-06-01 |
Family
ID=16503507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP58205227A Expired - Lifetime JPH0642158B2 (en) | 1983-11-01 | 1983-11-01 | Speech synthesizer |
Country Status (3)
| Country | Link |
|---|---|
| EP (1) | EP0144731B1 (en) |
| JP (1) | JPH0642158B2 (en) |
| DE (1) | DE3473956D1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02141054A (en) * | 1988-11-21 | 1990-05-30 | Nec Home Electron Ltd | Terminal equipment for personal computer communication |
| JP3070127B2 (en) * | 1991-05-07 | 2000-07-24 | 株式会社明電舎 | Accent component control method of speech synthesizer |
| DE19610019C2 (en) | 1996-03-14 | 1999-10-28 | Data Software Gmbh G | Digital speech synthesis process |
| JP4265501B2 (en) | 2004-07-15 | 2009-05-20 | ヤマハ株式会社 | Speech synthesis apparatus and program |
| JP5782751B2 (en) * | 2011-03-07 | 2015-09-24 | ヤマハ株式会社 | Speech synthesizer |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE2531006A1 (en) * | 1975-07-11 | 1977-01-27 | Deutsche Bundespost | Speech synthesis system from diphthongs and phonemes - uses time limit for stored diphthongs and their double application |
| JPS5331561A (en) * | 1976-09-04 | 1978-03-24 | Mitsukawa Shiyouichi | Method of manufacturing ssshaped springs |
| DE3105518A1 (en) * | 1981-02-11 | 1982-08-19 | Heinrich-Hertz-Institut für Nachrichtentechnik Berlin GmbH, 1000 Berlin | METHOD FOR SYNTHESIS OF LANGUAGE WITH UNLIMITED VOCUS, AND CIRCUIT ARRANGEMENT FOR IMPLEMENTING THE METHOD |
| JPS6017120B2 (en) * | 1981-05-29 | 1985-05-01 | 松下電器産業株式会社 | Phoneme piece-based speech synthesis method |
| JPS5868099A (en) * | 1981-10-19 | 1983-04-22 | 富士通株式会社 | Voice synthesizer |
| US4601052A (en) * | 1981-12-17 | 1986-07-15 | Matsushita Electric Industrial Co., Ltd. | Voice analysis composing method |
| NL8200726A (en) * | 1982-02-24 | 1983-09-16 | Philips Nv | DEVICE FOR GENERATING THE AUDITIVE INFORMATION FROM A COLLECTION OF CHARACTERS. |
| JPS5972494A (en) * | 1982-10-19 | 1984-04-24 | 株式会社東芝 | Rule snthesization system |
-
1983
- 1983-11-01 JP JP58205227A patent/JPH0642158B2/en not_active Expired - Lifetime
-
1984
- 1984-11-02 EP EP84113186A patent/EP0144731B1/en not_active Expired
- 1984-11-02 DE DE8484113186T patent/DE3473956D1/en not_active Expired
Also Published As
| Publication number | Publication date |
|---|---|
| DE3473956D1 (en) | 1988-10-13 |
| JPS6097396A (en) | 1985-05-31 |
| EP0144731A2 (en) | 1985-06-19 |
| EP0144731A3 (en) | 1985-07-03 |
| EP0144731B1 (en) | 1988-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3408477B2 (en) | Semisyllable-coupled formant-based speech synthesizer with independent crossfading in filter parameters and source domain | |
| US5400434A (en) | Voice source for synthetic speech system | |
| JPS62160495A (en) | Voice synthesization system | |
| JPH031200A (en) | Regulation type voice synthesizing device | |
| JP3732793B2 (en) | Speech synthesis method, speech synthesis apparatus, and recording medium | |
| JPH0642158B2 (en) | Speech synthesizer | |
| JP4510631B2 (en) | Speech synthesis using concatenation of speech waveforms. | |
| JP5175422B2 (en) | Method for controlling time width in speech synthesis | |
| JP2577372B2 (en) | Speech synthesis apparatus and method | |
| JPS5972494A (en) | Rule snthesization system | |
| JP3089940B2 (en) | Speech synthesizer | |
| JP2586040B2 (en) | Voice editing and synthesis device | |
| JPS5880699A (en) | Voice synthesizing system | |
| JP2573586B2 (en) | Rule-based speech synthesizer | |
| JP2573585B2 (en) | Speech spectrum pattern generator | |
| JPH0572599B2 (en) | ||
| JPS6295595A (en) | Voice response system | |
| JPS60113299A (en) | Voice synthesizer | |
| JPH09325788A (en) | Device and method for voice synthesis | |
| JP2001166787A (en) | Voice synthesizer and natural language processing method | |
| JPH0553595A (en) | Speech synthesizing device | |
| JPS58168096A (en) | Multi-language voice synthesizer | |
| JPS63285597A (en) | Phoneme connection type parameter rule synthesization system | |
| JPH11224096A (en) | Method and device for speech synthesis | |
| JPH06138894A (en) | Device and method for voice synthesis |