公开(公告)号：EP0829849A2

公开(公告)日：1998-03-18

申请号：EP97115693.0

申请日：1997-09-10

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Tanaka, Kimihito ,  Abe, Masanobu

IPC分类号： G10L5/04

CPC分类号： G10L13/07 ,  G10L13/04

摘要： Data in the same range of the fundamental frequency F 0 as speech segments are used as a learning data to prepare a reference codebook CB M for a spectrum envelope. The same learning data for a higher range than F 0 and the same learning data for a lower range are subject to a linear stretch matching with respect to the learning data for the range F 0 . For each vector code in the reference codebook CB M , the spectrum envelope is clustered to prepare a high range codebook CB H and a low range codebook CB L . The spectrum envelope of input speech segments are fuzzy vector quantized (S402) with the reference codebook, and depending on the synthesized F 0 , either one of high, middle and low codebooks is selected. The selected codebook is used to decode the fuzzy vector quantized code, and the decoded output is subject to the inverse FFT. Alternatively, codebooks CM MH and CB ML each comprising differential vectors for corresponding code vectors between CB M and CB H and between CB M and CB L are prepared. The quantized code is decoded using either CB MH or CB ML , and the decoded differential vector is stretched in accordance with a difference in the fundamental frequency between the synthesized speech and the original speech for CB M . The stretched differential vector is added the code vector which was used for the fuzzy vector quantization.

摘要翻译： 在相同范围内的基本频率F 0为语音段的数据被用作学习数据以制备参考码本CBM对频谱包络。 对于较高范围比F0和较低范围相同的学习数据同样的学习数据都受到相对于该范围F0的学习数据相匹配的线性伸展。 对于在参考码本CBM每个矢量代码，频谱包络进行聚类以制备高范围代码簿CBH和低范围代码簿CBL。 输入的语音段的频谱包络进行量化（S402）与所述参考码本模糊向量，并且根据合成F0，无论高，中，低的一个码本被选择。 所选择的代码簿用于解码的模糊向量量化码，并且将解码的输出是受逆FFT。 可替换地，代码簿和CMMH CBML对应CBM和CBH CBM和CBL之间之间的码矢量的每个包括差分矢量制备。 量化的代码是使用任一CBMH或CBML解码，并且将解码的差分矢量被拉伸在雅舞蹈与合成语音与原始语音为CBM之间的基本频率的差。 拉伸的差矢量加到其用于模糊矢量量化的代码矢量。

公开(公告)号：EP0829849B1

公开(公告)日：2003-08-06

申请号：EP97115693.0

申请日：1997-09-10

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Tanaka, Kimihito ,  Abe, Masanobu

IPC分类号： G10L13/02

CPC分类号： G10L13/07 ,  G10L13/04

公开(公告)号：EP0829849A3

公开(公告)日：1998-12-23

申请号：EP97115693.0

申请日：1997-09-10

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Tanaka, Kimihito ,  Abe, Masanobu

IPC分类号： G10L5/04

CPC分类号： G10L13/07 ,  G10L13/04

摘要： Data in the same range of the fundamental frequency F 0 as speech segments are used as a learning data to prepare a reference codebook CB M for a spectrum envelope. The same learning data for a higher range than F 0 and the same learning data for a lower range are subject to a linear stretch matching with respect to the learning data for the range F 0 . For each vector code in the reference codebook CB M , the spectrum envelope is clustered to prepare a high range codebook CB H and a low range codebook CB L . The spectrum envelope of input speech segments are fuzzy vector quantized (S402) with the reference codebook, and depending on the synthesized F 0 , either one of high, middle and low codebooks is selected. The selected codebook is used to decode the fuzzy vector quantized code, and the decoded output is subject to the inverse FFT. Alternatively, codebooks CM MH and CB ML each comprising differential vectors for corresponding code vectors between CB M and CB H and between CB M and CB L are prepared. The quantized code is decoded using either CB MH or CB ML , and the decoded differential vector is stretched in accordance with a difference in the fundamental frequency between the synthesized speech and the original speech for CB M . The stretched differential vector is added the code vector which was used for the fuzzy vector quantization.

公开(公告)号：EP0831460A3

公开(公告)日：1998-11-25

申请号：EP97116540.2

申请日：1997-09-23

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Abe, Masanobu

IPC分类号： G10L5/04

CPC分类号： G10L13/10

摘要： In a method and apparatus which use actual speech as auxiliary information and synthesize speech by speech synthesis by rule, prosodic information for a phoneme sequence of each word of a word sequence obtained by an analysis of an input text is set by referring to a word dictionary and a speech waveform sequence is obtained from the phoneme sequence of each word by referring to a speech waveform dictionary. On the other hand, prosodic information is extracted from input actual speech and either one of the set prosodic information and the extracted prosodic information is selected and the selected prosodic information is used to control the speech waveform sequence to create synthesized speech.

公开(公告)号：EP0831460B1

公开(公告)日：2003-02-26

申请号：EP97116540.2

申请日：1997-09-23

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Abe, Masanobu

IPC分类号： G10L13/08

CPC分类号： G10L13/10

公开(公告)号：EP0831460A2

公开(公告)日：1998-03-25

申请号：EP97116540.2

申请日：1997-09-23

申请人： NIPPON TELEGRAPH AND TELEPHONE CORPORATION

发明人： Abe, Masanobu

IPC分类号： G10L5/04

CPC分类号： G10L13/10

摘要： In a method and apparatus which use actual speech as auxiliary information and synthesize speech by speech synthesis by rule, prosodic information for a phoneme sequence of each word of a word sequence obtained by an analysis of an input text is set by referring to a word dictionary and a speech waveform sequence is obtained from the phoneme sequence of each word by referring to a speech waveform dictionary. On the other hand, prosodic information is extracted from input actual speech and either one of the set prosodic information and the extracted prosodic information is selected and the selected prosodic information is used to control the speech waveform sequence to create synthesized speech.

摘要翻译： 在使用实际语音作为辅助信息并通过规则进行语音合成来合成语音的方法和装置中，通过参考单词词典来设置通过分析输入文本而获得的单词序列的每个单词的音素序列的韵律信息 并且通过参考语音波形字典从每个单词的音素序列获得语音波形序列。 另一方面，从输入的实际语音中提取韵律信息，并且选择所设置的韵律信息和所提取的韵律信息中的任一个，并使用所选择的韵律信息来控制所述语音波形序列以创建合成语音。