Mitsubishi Denki Kabushiki Kaisha

Kunio Nakajima

Hirohisa Tasaki

Koichi Shiraki

Katsushi Seza

Shinobu Mizuta

The encoder unit of the vocoder device includes the AR code-book, the MA code-book, and the voice source code-book storing code words each corresponding to a set of the AR, the MA, and the voice source parameters, respectively, which are obtained beforehand by means of the 'analysis by synthesis' of a multitude of speech waveform examples and then clustering the resulting respective parameters. The AR preliminary selector, the MA preliminary selector, and the voice source preliminary selector select from respective code-books a predetermined finite number of code words approximating the input speech signal, and in synchronism with the voice source position detected by the voice source position detector the speech synthesizer synthesizes a number of synthesized speech waveforms corresponding to the combinations of the selected AR, MA, and voice source parameters. Comparing the synthesized speech waveforms with the current input speech signal waveform, the optimal code word selector selects the combination of the AR, the MA, and the voice source code words having a minimum distance to the input speech signal waveform.

Code-book driven vocoder device with voice source generator

Disclosed is an Hidden Markov Model (HMM) training apparatus in which a capacity for discriminating between models is taken into consideration so as to allow a high level of recognition accuracy to be obtained. A probability of a vector sequence appearing from HMMs is computed with respect to an input vector and continuous mixture density HMMs. Through this computation, the nearest different-category HMM, with which the maximum probability is obtained and which belongs to a category different from that of a training vector sequence of a known category, is selected. The respective central vectors of continuous densities constituting the output probability densities of the same-category HMM belonging to the same category as that of the training vector sequence and the nearest different-category HMM are moved on the basis of the vector sequence.

Pattern representation model training apparatus

A speech coder utilizing previously stored sound source vectors to generate synthetic speech, a distortion computing circuit for computing a distortion of synthetic speech from input speech and a selection circuit for selecting the sound source vector that provides minimum distortion. Sound source vectors are stored within a plurality of reduced size code books rather than a single larger code book. A vector adder adds the sound source vectors respectively output from each of the reduced code books thereby generating a single sound source vector for comparison with the input speech. The distortion circuit computes the distortion for this sound source vector by analyzing the sound source vectors respectively output from each of the reduced code books in addition to the sound source vector output from the vector adder. The computational complexity required to determine the distortion is greatly reduced from the complexity required if a single larger code book of sound source vectors is utilized.

Signal processor for analyzing distortion of speech signals

In a quantizer of the invention, a quantization distortion in the case where the optimum vector code word was used and a quantization distortion in the case where the optimum predictive error vector code word was used, are compared by comparing means, thereby selecting which one of the conventional vector quantization and the predictive error vector quantization should be executed for the present input vector. On the other hand, an inverse-quantizer selects which one of the inverse-quantizations should be executed on the basis of the input code number from the vector code. According to the invention, further, the quantizer and inverse-quantizer have the same analyzing means and converting means and execute the adaptation of the code tables by using the quantization results. Thus, the adaptation of the code table of the quantizer and inverse-quantizer can be accomplished without, particularly, adding the information transmitting process for adaptation. On the other hand, by executing the quantization by selectively using the predictive error vector quantizing means, the accuracy of the quantization result is improved and the adaptation of good code tables is realized.

Growing community of inventors

Kanagawa, Japan

Kunio Nakajima