JPS58107597A - Voice synthesizer - Google Patents
Voice synthesizerInfo
- Publication number
- JPS58107597A JPS58107597A JP56206609A JP20660981A JPS58107597A JP S58107597 A JPS58107597 A JP S58107597A JP 56206609 A JP56206609 A JP 56206609A JP 20660981 A JP20660981 A JP 20660981A JP S58107597 A JPS58107597 A JP S58107597A
- Authority
- JP
- Japan
- Prior art keywords
- chain
- phoneme
- parameter
- vowel
- phonetic
- 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.)
- Granted
Links
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Reciprocating, Oscillating Or Vibrating Motors (AREA)
Abstract
(57)【要約】本公報は電子出願前の出願データであるた
め要約のデータは記録されません。(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.
Description
【発明の詳細な説明】
本発明社音韻連鎖群によシ任意の語いの音声の生成を行
なう音声合成装置に関するものであ例えば時間波形その
ものＸあるいはホルマン周波数のよう表圧縮パラメータ
を線型又祉指数的に補間して合成波形を生成する方法が
行なわれていた。しかし、この方法で合成された音声は
、母音、子音間の音韻０つながシの点で自然性が劣石欠
点がある。この欠点を解消して合成音声の品質を改善す
るためには母音・子音・母音（ＶＯＶ）の音韻連鎖群を
用いる方法もあるが、この方法は大容量のメモリを必要
とするため装置が高価になる。本発明は子音・母音（Ｏ
ｖ）音韻連鎖群と母音群とよシなる小容量のメモリによ
り前記ＶＯＶ音韻連鎖群による合成音声と同等の品質を
有する合成音声なうるととの可能な音声合成装置を提供
することを目的とするものである。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a speech synthesis device that generates speech of arbitrary words using a phonological chain group. A method of generating a composite waveform by exponential interpolation has been used. However, the speech synthesized by this method has a disadvantage in that it lacks naturalness in that there are no phonetic connections between vowels and consonants. In order to overcome this drawback and improve the quality of synthesized speech, there is a method that uses vowel-consonant-vowel (VOV) phonological chains, but this method requires a large amount of memory and is expensive to use. become. The present invention focuses on consonants and vowels (O
v) It is an object of the present invention to provide a speech synthesis device capable of generating synthesized speech having the same quality as the synthesized speech using the VOV phonological chain group using a small memory capacity that includes a phonological chain group and a vowel group. It is something to do.
本発明は音韻連鎖データを圧縮パラメータ列として蓄積
する蓄積手段と、音声出力すべき音韻連鎖基を受信する
受信手段と、この受信手段により見られた音韻連鎖基に
対応する前記蓄積手段に蓄積され九圧縮パラメータを１
次読み出し、かつ練圧縮パラメータによシ音韻連鎖デー
タを合成する合成手段と、音韻連鎖Ｘと音韻連鎖１とが
連続する音韻連鎖である場合に、前記音韻連鎖！内の最
後部の母音パラメータ列Ｉと前記音韻連鎖！内の最前部
の母音パラメータＶＩと０２組の母音パラメータを用い
て前記音韻連鎖Ｘ内置後部の母音部出力終了後、前記パ
ラメータＶＸから前記パラメータＶＹＫ滑らかに推移す
る適蟲な時間長の音−補間データを合成する補間合成手
段とを備えていることを特徴とする音声合成装置である
。以下圧縮パラメータとしてネル１ント馬波数を用いた
実施例を第１図および籐２図について説明する。The present invention provides a storage means for storing phoneme chain data as a compressed parameter string, a reception means for receiving phoneme chain bases to be outputted as speech, and a storage means for storing phoneme chain bases corresponding to the phoneme chain bases seen by the receiving means. 9 compression parameters 1
Next, when the phoneme chain X and the phoneme chain 1 are consecutive phoneme chains, the synthesis means for synthesizing the phoneme chain data according to the refinement compression parameters, the phoneme chain! The vowel parameter sequence I at the end of and the phonological chain! After the output of the vowel part at the end of the phonological chain The present invention is a speech synthesis device characterized by comprising an interpolation synthesis means for synthesizing data. Hereinafter, an embodiment using the Nel and H wave numbers as compression parameters will be described with reference to FIGS. 1 and 2.
第１図において、受信部１に例えばＡＵ 。In FIG. 1, the receiving unit 1 includes, for example, an AU.
ＪＳ／Ｇ信号が受信されると２個の音韻連鎖基ＡＵ／、
／Ｊｉ／が合成部Ｓに送られる。合成部５は音韻連鎖基
／ｋＵ／に対応する圧縮パラメータ（ホルマント周波数
）をメモリ２から順次に読み出して合成する。ＡＴＥ／
の出力終了時に、補−関合成部４ ｉｊ ／ＫＴＪ／の
母音／Ｕ／の最後尾の圧縮パラメータとＡｉ／の母音Ａ
１０最前部の圧縮パラメータとをメモリよシ読み出して
第２図に示すように、線型補間をして出力する。補間時
間長は先行母音ＡＪ／の半分以下が１１ましい。When the JS/G signal is received, two phonetic chain bases AU/,
/Ji/ is sent to the synthesis section S. The synthesis unit 5 sequentially reads compression parameters (formant frequencies) corresponding to the phonetic chain base /kU/ from the memory 2 and synthesizes them. ATE/
At the end of the output, the complement synthesis unit 4 extracts the last compression parameter of the vowel /U/ of ij /KTJ/ and the vowel A of Ai/.
10 compression parameters at the front end are read out from the memory, subjected to linear interpolation, and output as shown in FIG. It is preferable that the interpolation time length be less than half of the preceding vowel AJ/.
補間脅威が終了すると合成部祉音韻連鎖ＡｉＺを上述の
ような方法で出力する０本発明は以上の作用を有するか
ら冒頭で述べ九本発明の所期の目的を達成する効果を有
する。When the interpolation threat is completed, the synthesis section outputs the phonological chain AiZ in the above-described manner.Since the present invention has the above-described operation, it has the effect of achieving the intended purpose of the present invention as stated at the beginning.
第１［ｌ：本発明の装置の構成を示すプロッタ図第２図
：本発明の作用説明図1 [l: Plotter diagram showing the configuration of the apparatus of the present invention Figure 2: Diagram explaining the operation of the present invention
Claims (1)
手段と、音声出力すべき音韻連鎖基を受信する受信手段
と、この受信手段によシ見られた音韻連鎖基に対応する
前記蓄積手段に蓄積された圧縮パラメータを順次読み出
し、かり練圧・縮パラメータにより音韻連鎖データを合
成する合成手段と、音韻連鎖Ｘと音韻連鎖！とが連続す
る音韻連鎖である場合に、前記音韻連鎖Ｘ内の鐘後部の
母音パラメータＶＸと前記音韻連鎖！内の最前部の母音
パラメータＶＹとの２組の母音パラメータを用いて前記
音韻連鎖Ｘ内 。 最優部の母音部出力終了後、前記パラメータＶ！から前
記パラメータＶ！に滑らかに推移する適当な時間長の音
韻補間データを合成する補間合成手段とを備えているこ
とを特徴とする音声合成装置[Scope of Claims] Accumulating means for accumulating phoneme chain data as a compressed parameter string; receiving means for receiving a phoneme chain base to be outputted as speech; A synthesis means that sequentially reads out the compression parameters stored in the storage means and synthesizes phoneme chain data using the compression/compression parameters, and phoneme chain X and phoneme chain! is a continuous phonetic chain, then the vowel parameter VX at the end of the bell in the phonetic chain X and the phonetic chain! In the phonological chain After outputting the vowel part of the most dominant part, the parameter V! From the parameter V! and an interpolation synthesis means for synthesizing phoneme interpolation data of an appropriate length that smoothly changes over time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP56206609A JPS5951000B2 (en) | 1981-12-21 | 1981-12-21 | speech synthesizer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP56206609A JPS5951000B2 (en) | 1981-12-21 | 1981-12-21 | speech synthesizer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS58107597A true JPS58107597A (en) | 1983-06-27 |
| JPS5951000B2 JPS5951000B2 (en) | 1984-12-11 |
Family
ID=16526213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP56206609A Expired JPS5951000B2 (en) | 1981-12-21 | 1981-12-21 | speech synthesizer |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS5951000B2 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS61106399A (en) * | 1984-10-30 | 1986-05-24 | 東北電力株式会社 | Height service car |
-
1981
- 1981-12-21 JP JP56206609A patent/JPS5951000B2/en not_active Expired
Also Published As
| Publication number | Publication date |
|---|---|
| JPS5951000B2 (en) | 1984-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3294604B2 (en) | Processor for speech synthesis by adding and superimposing waveforms | |
| US4220819A (en) | Residual excited predictive speech coding system | |
| US3995116A (en) | Emphasis controlled speech synthesizer | |
| JPH0122636B2 (en) | ||
| JPS58107597A (en) | Voice synthesizer | |
| JPH1078791A (en) | Pitch converter | |
| JP3059751B2 (en) | Residual driven speech synthesizer | |
| JPH08248993A (en) | Controlling method of phoneme time length | |
| JPH0258640B2 (en) | ||
| JPS6097396A (en) | Voice synthesizer | |
| JPS61278900A (en) | Voice synthesizer | |
| JPS5880699A (en) | Voice synthesizing system | |
| JPS60153099A (en) | Rule type voice synthesizer | |
| JPS58168097A (en) | Voice synthesizer | |
| JPH03160500A (en) | Speech synthesizer | |
| JPS60113299A (en) | Voice synthesizer | |
| JPS58129500A (en) | Singing voice synthesizer | |
| JPS5915299A (en) | Voice analyzer | |
| JPS58134697A (en) | Waveform editting type voice synthesizer | |
| JPH01118200A (en) | Voice synthesization system | |
| JPS63210900A (en) | Voice synthesizer | |
| JPS6432299A (en) | Unit voice editing type rule synthesizer | |
| JPH0464080B2 (en) | ||
| JPS5876900A (en) | Voice synthesization | |
| JPS5868799A (en) | Voice synthesizer |