Speech Analysis-synthesis System Research Articles

Speech analysis-synthesis system using genetic algorithm and Fujisaki model and its application to coarticulation

Speech analysis-synthesis system using genetic algorithm and Fujisaki model and its application to coarticulation

Speech quality evaluation for different pitch detection algorithms in LPC speech analysis–synthesis system

In this paper, the performance comparison of three pitch detection algorithms (PDAs) has been presented by implementing them in a LPC based speech analysis-synthesis system. The PDAs considered for comparison is based on three paradigms. The paradigms are weighted autocorrelation function (WACF), Empirical Mode Decomposition based autocorrelation function (EMD-ACF) and Empirical Mode Decomposition based average magnitude difference function (EMD-AMDF). Speech quality measurement is an important and essential task to ensure and maintain the quality of services for speech processing applications like modern telecommunication. Hence, the performance of these methods has been compared through the output speech quality using objective (perceptual evaluation of speech quality test) and subjective quality assessment (Mean Opinion Score test, diagnostic rhyme test and synthesized speech waveforms). The results show that the speech quality for the EMD-ACF and EMD-AMDF based PDA’s are better than that for WACF based PDA. The works presented in this paper is beneficial to telecommunication and speech recognition research group.

Real‐time implementation and performance evaluation of speech classifiers in speech analysis‐synthesis

In this work, six voiced/unvoiced speech classifiers based on the autocorrelation function (ACF), average magnitude difference function (AMDF), cepstrum, weighted ACF (WACF), zero crossing rate and energy of the signal (ZCR-E), and neural networks (NNs) have been simulated and implemented in real time using the TMS320C6713 DSP starter kit. These speech classifiers have been integrated into a linear-predictive-coding-based speech analysis-synthesis system and their performance has been compared in terms of the percentage of the voiced/unvoiced classification accuracy, speech quality, and computation time. The results of the percentage of the voiced/unvoiced classification accuracy and speech quality show that the NN-based speech classifier performs better than the ACF-, AMDF-, cepstrum-, WACF- and ZCR-E-based speech classifiers for both clean and noisy environments. The computation time results show that the AMDF-based speech classifier is computationally simple, and thus its computation time is less than that of other speech classifiers, while that of the NN-based speech classifier is greater compared with other classifiers.

Speech analysis-synthesis system using principal components of vowel spectra

We have studied an effective method using principal components spanning a feature space of isolated vowels. A covariance matrix is calculated from many log-amplitude spectra of isolated vowels uttered by a speaker. An eigen equation of the covariance matrix is solved. The resulting eigenvectors are called principal vectors. In the analysis system, log-amplitude spectrum for each frame of a word uttered by the same speaker is transformed to the components on the principal vectors. In the synthesis system, a log-amplitude spectrum is reconstructed using the components on the principal vectors with the largest eigenvalues and the spoken word is synthesized using the LMA filter. We draw the distribution chart of the first and the second principal components extracted from Japanese vowel data. This figure was very similar to the F1—F2 distribution of vowels and so to the vowel classification map in coordinate axes of the degree of constriction and the tongue hump position. The Listening tests showed that the q...

Performance Evaluation of Novel AMDF-Based Pitch Detection Scheme

A novel average magnitude difference function (AMDF)-based pitch detection scheme (PDS) is proposed to achieve better performance in speech quality. A performance evaluation of the proposed PDS is carried out through both a simulation and a real-time implementation of a speech analysis-synthesis system. The parameters used to compare the performance of the proposed PDS with that of PDSs that are based on either a cepstrum, an autocorrelation function (ACF), an AMDF, or circular AMDF (CAMDF) methods are as follows: percentage gross pitch error (%GPE); a subjective listening test; an objective speech quality assessment; a speech intelligibility test; a synthesized speech waveform; computation time; and memory consumption. The proposed PDS results in lower %GPE and better synthesized speech quality and intelligibility for different speech signals as compared to the cepstrum-, ACF-, AMDF-, and CAMDF-based PDSs. The computational time of the proposed PDS is also less than that for the cepstrum-, ACF-, and CAMDF-based PDSs. Moreover, the total memory consumed by the proposed PDS is less than that for the ACF- and cepstrum-based PDSs.

Performance evaluation of a ACF-AMDF based pitch detection scheme in real-time

In this paper, a complete speech analysis-synthesis system for a pitch detection algorithm based on short-time autocorrelation function (ACF) and average magnitude difference function (AMDF) has been implemented in real-time using TMS320C6713 DSK. Performance of this system has been compared with the analysis-synthesis system that use autocorrelation, cepstrum and wavelet based pitch detection method in terms of synthesized speech quality, Diagnostic Rhyme Test, execution time and memory consumption. Results show that this method of pitch detection scheme is the best in terms of speech quality, execution time and memory consumption. Moreover the synthesized speech using ACF-AMDF method of pitch detection is more intelligible compared to other pitch detection algorithm.

Time-frequency voiced and unvoiced excitation models for harmonic speech systems

Time-frequency voiced and unvoiced models are proposed for the excitation of a harmonic autoregressive wideband speech analysis-synthesis system. The time-frequency voiced excitation (TFVEX) model has low time resolution defined by the concentration of the excitation signal distribution in the modulation domain while the time-frequency unvoiced excitation (TFUNEX) model has cycle time discrimination with lower amplitude resolution and while the frequency resolution for both models is an octave. The speech reconstructed by the compound TFUVEX unvoiced-voiced model is rated above the speech degraded by a modulated noise reference unit (MNRU) at 25 dB in listening tests while yielding a parametric compression of over ten times.

Time-frequency voiced and unvoiced excitation models for harmonic speech systems DOI: 10.14209/jcis.2014.6

Time-frequency voiced and unvoiced models are proposed for the excitation of a harmonic autoregressive wideband speech analysis-synthesis system. The time-frequency voiced excitation (TFVEX) model has low time resolution defined by the concentration of the excitation signal distribution in the modulation domain while the time-frequency unvoiced excitation (TFUNEX) model has cycle time discrimination with lower amplitude resolution and while the frequency resolution for both models is an octave. The speech reconstructed by the compound TFUVEX unvoiced-voiced model is rated above the speech degraded by a modulated noise reference unit (MNRU) at 25 dB in listening tests while yielding a parametric compression of over ten times.

Performance evaluation of a wavelet-based pitch detection scheme

Performance evaluation, in a complete speech analysis-synthesis system, has been carried out for a wavelet-based pitch detection scheme that has been reported earlier. Speech quality, time for computation and memory consumption (for real-time implementation) are the parameters that have been considered while comparing this system with analysis-synthesis systems that use pitch detection based on autocorrelation and cepstral analysis. Results for different speech signals show that autocorrelation-based pitch detection scheme is the best in terms of speech quality and memory consumption while wavelet-based pitch detection stands in between the other two methods.

Synthesis fidelity and time-varying spectral change in vowels

Recent studies have shown that synthesized versions of American English vowels are less accurately identified when the natural time-varying spectral changes are eliminated by holding the formant frequencies constant over the duration of the vowel. A limitation of these experiments has been that vowels produced by formant synthesis are generally less accurately identified than the natural vowels after which they are modeled. To overcome this limitation, a high-quality speech analysis-synthesis system (STRAIGHT) was used to synthesize versions of 12 American English vowels spoken by adults and children. Vowels synthesized with STRAIGHT were identified as accurately as the natural versions, in contrast with previous results from our laboratory showing identification rates 9%-12% lower for the same vowels synthesized using the cascade formant model. Consistent with earlier studies, identification accuracy was not reduced when the fundamental frequency was held constant across the vowel. However, elimination of time-varying changes in the spectral envelope using STRAIGHT led to a greater reduction in accuracy (23%) than was previously found with cascade formant synthesis (11%). A statistical pattern recognition model, applied to acoustic measurements of the natural and synthesized vowels, predicted both the higher identification accuracy for vowels synthesized using STRAIGHT compared to formant synthesis, and the greater effects of holding the formant frequencies constant over time with STRAIGHT synthesis. Taken together, the experiment and modeling results suggest that formant estimation errors and incorrect rendering of spectral and temporal cues by cascade formant synthesis contribute to lower identification accuracy and underestimation of the role of time-varying spectral change in vowels.

Vector quantizing apparatus and speech analysis-synthesis system using the apparatus

A vector quantizing apparatus having a general vector quantization circuit, and a storage means for storing at least one frame of data as the result of comparison by a matching circuit is provided. Further, provided are a speech analysis-synthesis system having a spectral envelope generator for generating a spectral envelope which is so smooth that excessive beating is avoided, a spectral envelope vector converter for sampling the spectral envelope at equal intervals on mel-scale, a vector quantizer for quantizing vectors, and a spectral envelope reconstructor for reconstructing the spectral envelope by interpolation on the basis of combined parabolas.

Speech analysis–synthesis system

First Page

パワースペクトル包絡(PSE)音声分析・合成系

パワースペクトル包絡(PSE)音声分析・合成系

Low-dimensional representation of vowels based on all-pole modeling in the psychophysical domain

A novel speech analysis method which uses several established psychoacoustic concepts is applied to the analysis of vowels. This perceptually based linear predictive analysis (PLP) models the auditory spectrum by the spectrum of the low-order all-pole model. The auditory spectrum is derived from the speech waveform by critical-band filtering, equal-loudness curve pre-emphasis, and intensity-loudness root compression. We demonstrate through analysis of both natural and synthetic speech that psychoacoustic concepts of spectral auditory integration in vowel perception, namely the F1, F2′ concept of Carlson and Fant and the 3.5 Bark auditory integration concept of Chistovich, are well modeled by the PLP method. A complete speech analysis-synthesis system based on the PLP method is also described in the paper.

High quality system for speech transformations

A speech analysis synthesis system has been developed which is capable of independently modifying the excitation and the spectrum, and then reconstructing speech from the modified components. Since pitch extraction is not required, the reconstructed speech is more natural sounding than vocoded speech. At the core of the system is a phase vocoder. The excitation spectrum is obtained by dividing the magnitude spectrum by the vocal tract spectral model, and is then either duplicated or foreshortened to alter the number of harmonics present. Correct frequencies are restored by multiplying the unwrapped phase characteristics by the appropriate constant. The desired vocal tract shaping is then reintroduced, using interpolation between samples if necessary. The system can produce several potentially useful modifications to the voice. A male voice can be converted into a female‐like voice, and vice versa. Alternatively, the spectrum can be compressed by a factor of three or more, while the pitch remains unchanged; a modification potentially useful for persons with high frequency hearing loss. A third possibility is the restoration of helium speech. In addition, the parameters of the system could be encoded to realize a voice‐excited vocoder. A tape will be played illustrating the system's capabilities.

Speech synthesis by linear interpolation of spectral parameters between dyad boundaries.

A recent study [Olive and Spickenagel, J. Acoust. Soc. Am. 59, 993-996 (1976)] has shown that area parameters derived from linear prediction analysis can be linearly interpolated between dyad boundaries with very little distortion in the resultant synthesized speech. The success of area parameter interpolation raises a question: can other acoustic parameters, such as the power spectrum of the speech waveform, be similarly interpolated? The spectrum is of special interest because speech can be synthesized in real time from spectral parameters on a programmable digital filter. To study this question a speech analysis-synthesis system using spectral parameters (samples of power spectra at different frequencies) was simulated. These parameters were determined from the speech signal at every dyad boundary, and interpolated for intermediate values. Dyad boundaries (representing the limits of transition regions between phonemes) were determined manually. Informal listening tests comparing synthetic speech with and without linear interpolation showed slight degradation in the interpolated speech. This degradation is significantly reduced by using an additional point within the dyad boundaries for interpolation.

Linear prediction with a variable analysis frame size

This paper describes a speech analysis-synthesis system based on stationary linear prediction formulation. This system uses a variable analysis frame size concept. The k-parameters are used to represent the spectral information in the speech. The statistical and quantization properties of k-parameters are studied in detail. A method for calculating the analysis frame size based on energy and pitch period variations within a speech waveform has been developed. The speech analysis-synthesis system has been implemented on the computing facility of the Signal Processing Laboratory at Case Western Reserve University. Average data rates of 4800, 3600, and 2400 bits/s have been achieved on a limited speech data base of male speakers.

Formant tracking algorithm for a speech analysis-synthesis system

A formant tracking algorithm has been developed to provide control signals automatically for a parallel formant speech synthesizer. In voiced sounds, the algorithm first estimates positions of high-energy excitation in the sampled speech waveform to be analyzed. These positions generally correspond to the points of glottal closure, and the interval following each position should correspond to a closed glottis region. For each such position the speech samples in the following few milliseconds are used to calculate a log power spectrum. An analysis-by-synthesis procedure iteratively updates estimates of formant frequencies and amplitudes using spectral matching criteria. A similar spectral-matching procedure is used during unvoiced sounds, but time averaging is introduced as appropriate when deriving the spectra to be matched. To cope with situations when the formant structure is not clearly defined in the power spectrum the iteration is performed separately for several different allocations of formants to spectral peaks. A final choice of formant frequencies and amplitudes is made every 10 msec and depends on both goodness of spectral match and continuity of formant tracks. Demonstrations will be made of the speech quality achieved for a number of male talkers.

Perceptions of personality from speech: effects of manipulations of acoustical parameters.

A speech analysis-synthesis system was used to manipulate variance of fundamental frequency and a mechanical rate changer was used to manipulate speech rate. The synthesized and altered voices were tested for realism. Synthesized voices were mistaken for normal 50% to 58% of the time and rate-changed voices were mistaken for normal 78% of the time. Additional studies were conducted to test the effects of these acoustical manipulations on the adjective ratings judges made of speakers. Variance of intonation was increased and decreased by 50% for eight speakers. There was a significant trend for increased intonation to cause voices to be rated more “benevolent” by judges and decreased intonation to cause them to be rated less “benevolent.” In two additional studies, rate was decreased and increased by varying amounts. Slowing the voices caused them to be rated less “competent.” Speeding the voices caused them to be rated less “benevolent.” Results were more consistent over speakers for rate manipulations than for intonation manipulations.

Design and simulation of a speech analysis-synthesis system based on short-time Fourier analysis

This paper discusses the theoretical basis for representation of a speech signal by its short-time Fourier transform. The results of the theoretical studies were used to design a speech analysis-synthesis system which was simulated on a general-purpose laboratory digital computer system. The simulation uses the fast Fourier transform in the analysis stage and specially designed finite duration impulse response filters in the synthesis stage. The results of both the theoretical and computational studies lead to an understanding of the effect of several design parameters and elucidate the design tradeoffs necessary to achieve moderate information rate reductions.