Pitch Synchronous Overlap And Add Research Articles

Age-Based Automatic Voice Conversion Using Blood Relation for Voice Impaired

The present work presents a statistical method to translate human voices across age groups, based on commonalities in voices of blood relations. The age-translated voices have been naturalized extracting the blood relation features e.g., pitch, duration, energy, using Mel Frequency Cepstrum Coefficients (MFCC), for social compatibility of the voice-impaired. The system has been demonstrated using standard English and an Indian language. The voice samples for resynthesis were derived from 12 families, with member ages ranging from 8–80 years. The voice-age translation, performed using the Pitch synchronous overlap and add (PSOLA) approach, by modulation of extracted voice features, was validated by perception test. The translated and resynthesized voices were correlated using Linde, Buzo, Gray (LBG), and Kekre’s Fast Codebook generation (KFCG) algorithms. For translated voice targets, a strong (θ >~93% and θ >~96%) correlation was found with blood relatives, whereas, a weak (θ

Synthesis of Emotional Speech by Prosody Modification of Vowel Segments of Neutral Speech

Background: Emotional speech synthesis is the process of synthesising emotions in a neutral speech – potentially generated by a text-to-speech system – to make an artificial humanmachine interaction human-like. It typically involves analysis and modification of speech parameters. Existing work on speech synthesis involving modification of prosody parameters does so at sentence, word, and syllable level. However, further fine-grained modification at vowel level has not been explored yet, thereby motivating our work. Objective: To explore prosody parameters at vowel level for emotion synthesis. Methods: Our work modifies prosody features (duration, pitch, and intensity) for emotion synthesis. Specifically, it modifies the duration parameter of vowel-like and pause regions and the pitch and intensity parameters of only vowel-like regions. The modification is gender specific using emotional speech templates stored in a database and done using Pitch Synchronous Overlap and Add (PSOLA) method. Results: Comparison was done with the existing work on prosody modification at sentence, word and syllable label on IITKGP-SEHSC database. Improvements of 8.14%, 13.56%, and 2.80% for emotions angry, happy, and fear respectively were obtained for the relative mean opinion score. This was due to: (1) prosody modification at vowel-level being more fine-grained than sentence, word, or syllable level and (2) prosody patterns not being generated for consonant regions because vocal cords do not vibrate during consonant production. Conclusion: Our proposed work shows that an emotional speech generated using prosody modification at vowel-level is more convincible than prosody modification at sentence, word and syllable level.

Structure of information technology for the removal of emotional signs from a speech signal

The goal of the article is to describe the methods and algorithms of information technology for removal of emotional signs from the speech signal which has been proposed by the authors. Methods of extracting and transforming emotional features, as well as methods of classification of speech signals obtained after transformation are described. The task of removing emotions will be considered successful if, after the transformations, the emotion classifier assigns a speech signal to a class of emotionless speech. The scheme for solving the problem consists of the following steps: 1) extracting informative features and establishing the boundaries of conversion; 2) transforming input data informative features; 3) checking the effectiveness of the transformation. The first algorithm used for transformation is the PSOLA (Pitch-Synchronous OverLap and Add) algorithm designed to change the pitch of the main tone by stretching and compressing the signal on the timeline. The PSOLA algorithm can be used to shift the pitch of voice, retaining the formant position, and thus the identity of the vowels. The basic idea is to extend the time on pitch marks, while the shape of the audio wave of the segment does not change. The second method of transformation, the normalization of the dynamic range of the logarithm of energy, reduces the distribution of energy due to different levels of background noise [2]. The input parameter for the energy normalization unit is the frame energy. Energy is replaced by the energy logarithm based on the similarity to the logarithmic perception of sound by the human auditory system, and also due to the stability of the logarithm to the sharp changes in the energy level. To test the effectiveness of the transformation, the task of recognizing emotions in a language requires the building of a classifier. The input data for classifying and recognizing were emotional expressions, each of which was represented by a set of signs: the frequency of the main tone, the energy and the frequency of crossing zero. Global signs of the minimum, the maximum and the average of each of the sets of values are the vector of expressions. For the classification a model of a mixture of normal sections was used. Using the constructed statistical model of the speech signal, the a posteriori probability of belonging of the utterance to the emotionally neutral class was calculated. The output speech signal meets the set criterion of a successful solution of the problem as formal (result of classification) and perceptual signs, which allows to consider the task of removing emotional features solved.

Categorical perception of lexical tones in native Mandarin-speaking listeners with sensorineural hearing loss

Background: Categorical perception (CP) of lexical tones was examined in normal hearing (NH) people, but it was unclear whether lexical tones can be perceived categorically in sensorineural hearing loss (SNHL) people.Objectives: To explore the characteristic of lexical tone perception in native Mandarin speakers with SNHL.Materials and methods: Three types of continuum (Tone1/Tone2, Tone1/Tone4 and Tone2/Tone3) were constructed and each of them includes 15 stimuli which were resynthesized by applying the pitch-synchronous overlap and add (PSOLA) method implemented in Praat to the same Mandarin syllable, /a/, with a high-level tone produced by a female speaker. Forty native Mandarin NH speakers and 23 native Mandarin speakers with mild to moderate SNHL were recruited. A two alternative-forced-choice identification task was used to acquire the tonal perceptual data.Results: All tone perception curves owns the characteristic of CP in SNHL subjects. All tone perception curves were S-shape in SNHL subjects same as those in NH subjects. No significant difference of each continuum was observed between SNHL and NH.Conclusions: CP of lexical tone perception could be observed in native Mandarin speakers with mild to moderate SNHL. The slight damage in the peripheral auditory system did not change characteristic of lexical tone perception.

The characteristic of lexical tone perception in native Mandarin speakers with mild and moderate sensorineural hearing loss

Objective:To explore the characteristic of lexical tone perception in native-Mandarin speakers with mild to moderate sensorineural hearing loss. Method:Three types of continuum(Tone1/tone2, tone1/tone4 and tone2/tone3) were constructed and each of them includes a 15 stimuli. All speech stimuli in these three continua were resynthesized by applying the Pitch-Synchronous OverLap and Add(PSOLA) method implemented in Praat to the same mandarin syllable, /a/, with a high level tone produced by a female native-mandarin speaker. Twenty-three native-mandarin sensory hearing loss people were recruited for this study and identification task was used to acquire the tonal perceptual data. SPSS 19.0 software was used to get the fitting curves of lexical tone perception. Result:①Each tone perception curves owns the characteristic of categorical perception, including an obvious boundary, a definite intersection and the abrupt performance change at intersection. The tone perception curves were S-shape in SNHL people which were same as that in normal hearing people.②No significant performance difference of each continuum was observed between SNHL and NH in this study(P=0.811 in T1/T2 continuum, P=0.528 in T1/T4 continuum, P=0.555 in T2/T3 continuum). Conclusion:There is the categorical perception of lexical tone identification in native-mandarin sensorineural hearing loss people. The slight damage in peripheral auditory system did not changed characteristic of lexical tone perception.

The characteristic of lexical tone perception in native-mandarin normal hearing people

Objective:The aim of this study is to explore the characteristic of lexical tone perception in nativemandarin normal hearing speakers.Method:Three types of continuum(Tone1/Tone2, Tone1/Tone4 and Tone2/Tone3) were constructed and each of them includes a 15 stimuli. All speech stimuli in these three continua were resynthesized by applying the PitchSynchronous OverLap and Add(PSOLA) method implemented in Praat to the same mandarin syllable, /a/, with a high level tone produced by a female native mandarin speaker. Forty native listeners of mandarin were recruited for this study and identification task was used to acquire the tonal perceptual data. SPSS 19.0 software was used to get the fitting curves of lexical tone perception. Result: The tone perception curves in nativemandarin normal hearing speakers were inverted Sshape. All the three tone perception curves showed steep upward or downward trend and obvious boundary.Each of tone perception curves has a definite intersection, and the tone perception performance has mutated at the intersection.Conclusion:There is the typically categorical perception of lexical tone identification in nativemandarin normal hearing people.

Perceptual versus cognitive speed in a time-compressed speech task

Time-compression retains the information-bearing spectral change present uncompressed speech, although at a rate that may outstrip cognitive processing speed. To compare the relative importance of perceptual and cognitive processing speed, we compared the understanding of (1) time-compressed stimuli expanded in time via gaps with (2) uncompressed stimuli where spectral change information was removed. We hypothesized that, despite the initial compression, the compressed and expanded stimuli would be more intelligible as it would retain relatively more information-bearing spectral change. Participants were somewhat older listeners (mid-1950s to mid-1960s) with normal hearing or mild hearing loss. Stimuli were spoken seven-digit strings time-compressed via pitch synchronous overlap and add (PSOLA) at three uniform compression ratios (2:1, 3:1, and 5:1). In gap insertion conditions, the total duration of the compressed stimuli was restored via introduction of periodic gaps. This produced signal-to-gap ratios ...

A Discrete Fractional Cosine Transform Based Speech Enhancement System Through Adaptive Kalman Filter Combined with Perceptual Weighting Filter with Pitch Synchronous Analysis

The speech enhancement plays a vital role commonly used in noisy environment to develop the performance of speech identification in mobile phones or in car navigation system. Thus the quality of the performance of the speech recognition is becoming worse due to the presence of noises in the surrounding. The objective is to increase the evident quality of the speech and to develop the transparency. Signal representation and enhancement in cosine transformation is observed to provide significant results. As an alternative of using DCT, a combination of conventional Discrete Cosine Transform (DCT) and Discrete Fourier Transform (DFT) which forms the another transform called as the Discrete Fractional Cosine Transform (DFrCT). The DFrCT have a free parameter, its fraction. In order to deal with the issue of frame to frame deviations of the Cosine Transformations, DFrCT is integrated with Pitch Synchronous Analysis (PSA). Also Pitch Synchronous OverLap and Add (PSOLA) method are used to enhance the performance of PSA. Moreover, in order to improve the noise minimization of the system, Improved Iterative Wiener Filtering approach called Adaptive Kalman Filter Combined with Perceptual Weighting Filter is used in this approach. This filter is used to eliminate the matrix operations, reduces both the calculation time and complexity. Thus, a novel DFrCT based speech enhancement using improved iterative filtering algorithm integrated with PSA is used in this approach

Simultaneous F 0-F 1 modifications of Arabic for the improvement of natural-sounding

Pitch (F 0) modification is one of the most important problems in the area of speech synthesis. Several techniques have been developed in the literature to achieve this goal. The main restrictions of these techniques are in the modification range and the synthesised speech quality, intelligibility and naturalness. The control of formants in a spoken language can significantly improve the naturalness of the synthesised speech. This improvement is mainly dependent on the control of the first formant (F 1). Inspired by this observation, this article proposes a new approach that modifies both F 0 and F 1 of Arabic voiced sounds in order to improve the naturalness of the pitch shifted speech. The developed strategy takes a parallel processing approach, in which the analysis segments are decomposed into sub-bands in the wavelet domain, modified in the desired sub-band by using a resampling technique and reconstructed without affecting the remained sub-bands. Pitch marking and voicing detection are performed in the frequency decomposition step based on the comparison of the multi-level approximation and detail signals. The performance of the proposed technique is evaluated by listening tests and compared to the pitch synchronous overlap and add (PSOLA) technique in the third approximation level. Experimental results have shown that the manipulation in the wavelet domain of F 0 in conjunction with F 1 guarantees natural-sounding of the synthesised speech compared to the classical pitch modification technique. This improvement was appropriate for high pitch modifications.

Voice attractiveness: Influence of stimulus duration and type

Voice attractiveness is a relatively new area of research. Some aspects of the methodology used in this domain deserve particular attention. Especially, the duration of voice samples is often neglected as a factor and happens to be manipulated without the perceptual consequences of these manipulations being known. Moreover, the type of voice stimulus varies from a single vowel to complex sentences. The aim of this experiment was to investigate the extent to which stimulus duration (nonmanipulated vs. normalized) and type (vowel vs. word) influence perceived voice attractiveness. Twenty-seven male and female raters made attractiveness judgments of 30 male and female voice samples. Voice samples included a single vowel /a/, a three-vowel series /i a o/, and the French word "bonjour" (i.e., "hello"). These samples were presented in three conditions: nonmanipulated, shortened, and lengthened duration. Duration manipulation was performed using the pitch synchronous overlap and add (PSOLA) algorithm implemented in Praat. Results for the effect of stimulus type showed that word length samples were more attractive to the opposite sex than vowels. Results for the effect of duration showed that the nonmanipulated sound sample duration was not predictive of perceived attractiveness. Duration manipulation, on the other hand, altered perceived attractiveness for the lengthening condition. In particular, there was a linear decrease in attractiveness as a function of modification percentage (especially for the word, as compared with the vowels). Recommendations for voice sample normalization with the PSOLA algorithm are thus to prefer shortening over lengthening and, if not possible, to limit the extent of duration manipulation-for example, by normalizing to the mean sample duration.

A pitch synchronous approach to design voice conversion system using source-filter correlation

We propose a pitch synchronous approach to design the voice conversion system taking into account the correlation between the excitation signal and vocal tract system characteristics of speech production mechanism. The glottal closure instants (GCIs) also known as epochs are used as anchor points for analysis and synthesis of the speech signal. The Gaussian mixture model (GMM) is considered to be the state-of-art method for vocal tract modification in a voice conversion framework. However, the GMM based models generate overly-smooth utterances and need to be tuned according to the amount of available training data. In this paper, we propose the support vector machine multi-regressor (M-SVR) based model that requires less tuning parameters to capture a mapping function between the vocal tract characteristics of the source and the target speaker. The prosodic features are modified using epoch based method and compared with the baseline pitch synchronous overlap and add (PSOLA) based method for pitch and time scale modification. The linear prediction residual (LP residual) signal corresponding to each frame of the converted vocal tract transfer function is selected from the target residual codebook using a modified cost function. The cost function is calculated based on mapped vocal tract transfer function and its dynamics along with minimum residual phase, pitch period and energy differences with the codebook entries. The LP residual signal corresponding to the target speaker is generated by concatenating the selected frame and its previous frame so as to retain the maximum information around the GCIs. The proposed system is also tested using GMM based model for vocal tract modification. The average mean opinion score (MOS) and ABX test results are 3.95 and 85 for GMM based system and 3.98 and 86 for the M-SVR based system respectively. The subjective and objective evaluation results suggest that the proposed M-SVR based model for vocal tract modification combined with modified residual selection and epoch based model for prosody modification can provide a good quality synthesized target output. The results also suggest that the proposed integrated system performs slightly better than the GMM based baseline system designed using either epoch based or PSOLA based model for prosody modification.

Pitch alteration technique in speech synthesis system

In the case of speech synthesis, the waveform coding method with its high quality is mainly used in synthesis by analysis. Because the parameters of this coding method are not classified as both the excitation and vocal tract parameters, it is difficult to apply the waveform coding method to synthesis by rule. Thus, in order to apply the waveform coding method to synthesis by rule, a pitch alteration is required for the prosody control. In the speech synthesis method by the conventional PSOLA (pitch synchronous overlap and add) technique, applying a symmetric window function to the asymmetric speech waveform, results in the energy unbalance phenomenon according to the degree of overlapped in the pitch interval adjustment. In this paper, to overcome the energy unbalance phenomenon, we proposed a new method that can convert the asymmetric waveform to a symmetric one by time-frequency conversion. As a result, we can obtain an average spectrum distortion ratio of 6.38% according to the pitch alteration ratio.

Psychoacoustical evaluation of PSOLA. II. Double-formant stimuli and the role of vocal perturbation.

This article presents the results of listening experiments and psychoacoustical modeling aimed at evaluating the pitch synchronous overlap-and-add (PSOLA) technique. This technique can be used for simultaneous modification of pitch and duration of natural speech, using simple and efficient time-domain operations on the speech waveform. The first set of experiments tested the ability of subjects to discriminate double-formant stimuli, modified in fundamental frequency using PSOLA, from unmodified stimuli. Of the potential auditory discrimination cues induced by PSOLA, cues from the first formant were found to generally dominate discrimination performance. In the second set of experiments the influence of vocal perturbation, i.e., jitter and shimmer, on discriminability of PSOLA-modified single-formant stimuli was determined. The data show that discriminability deteriorates at most modestly in the presence of jitter and shimmer. With the exception of a few conditions, the trends in these data could be replicated by either using a modulation-discrimination or an intensity-discrimination model, dependent on the formant frequency. As a baseline experiment detection thresholds for jitter and shimmer were measured. Thresholds for jitter could be replicated by using either the modulation-discrimination or the intensity-discrimination model, dependent on the (mean) fundamental frequency of stimuli. The thresholds for shimmer could be accurately predicted for stimuli with a 250-Hz fundamental, but less accurately in the case of a 100-Hz fundamental.

Psychoacoustical evaluation of the pitch-synchronous overlap-and-add speech-waveform manipulation technique using single-formant stimuli.

This article presents two experiments dealing with a psychoacoustical evaluation of the pitch-synchronous overlap-and-add (PSOLA) technique. This technique has been developed for modification of duration and fundamental frequency of speech and is based on simple waveform manipulations. Both experiments were aimed at deriving the sensitivity of the auditory system to the basic distortions introduced by PSOLA. In experiment I, manipulation of fundamental frequency was applied to synthetic single-formant stimuli under minimal stimulus uncertainty, level roving, and formant-frequency roving. In experiment II, the influence of the positioning of the so-called "pitch markers" was studied. Depending on the formant and fundamental frequency, experimental data could be described reasonably well by either a spectral intensity-discrimination model or a temporal model based on detecting changes in modulation of the output of a single auditory filter. Generally, the results were in line with psychoacoustical theory on the auditory processing of resolved and unresolved harmonics.

Voice transformation using PSOLA technique

In this contribution, a new system for voice conversion is described. The proposed architecture combines a PSOLA (Pitch Synchronous Overlap and Add)-derived synthesizer and a module for spectral transformation. The synthesizer based on the classical source-filter decomposition allows prosodic and spectral transformations to be performed independently. Prosodic modifications are applied on the excitation signal using the TD-PSOLA scheme; converted speech is then synthesized using the transformed spectral parameters. Two different approaches to derive spectral transformations, borrowed from the speech-recognition domain, are compared: Linear Multivariate Regression (LMR) and Dynamic Frequency Warping (DFW). Vector-quantization is carried out as a preliminary stage to render the spectral transformations dependent of the acoustical realization of sounds. A formal listening test shows that the synthesizer produces a satisfyingly natural “transformed” voice. LMR proves yet to allow a slightly better conversion than DFW. Still there is room for improvement in the spectral transformation stage.

Evaluation of speech synthesis techniques in a comprehension task

Six types of speech synthesis were evaluated for comprehensibility: standard linear predictive coding analysis/ resynthesis; pitch synchronous analysis/resynthesis; pitch synchronous multi-pulse analysis/resynthesis; and three PSOLA (pitch synchronous overlap-and-add) techniques. The relative comprehensibility of the synthesis types was tested by using the synthesised speech to convey information that subjects needed in order to perform a diagram-based multiple-choice task.