Speaker-independent Automatic Speech Recognition Research Articles

Maxillectomy patients' speech and performance of contemporary speaker-independent automatic speech recognition platforms in Japanese.

Automatic speech recognition (ASR) can potentially help older adults and people with disabilities reduce their dependence on others and increase their participation in society. However, maxillectomy patients with reduced speech intelligibility may encounter some problems using such technologies. To investigate the accuracy of three commonly used ASR platforms when used by Japanese maxillectomy patients with and without their obturator placed. Speech samples were obtained from 29 maxillectomy patients with and without their obturator and 17 healthy volunteers. The samples were input into three speaker-independent speech recognition platforms and the transcribed text was compared with the original text to calculate the syllable error rate (SER). All participants also completed a conventional speech intelligibility test to grade their speech using Taguchi's method. A comprehensive articulation assessment of patients without their obturator was also performed. Significant differences in SER were observed between healthy and maxillectomy groups. Maxillectomy patients with an obturator showed a significant negative correlation between speech intelligibility scores and SER. However, for those without an obturator, no significant correlations were observed. Furthermore, for maxillectomy patients without an obturator, significant differences were found between syllables grouped by vowels. Syllables containing /i/, /u/ and /e/ exhibited higher error rates compared to those containing /a/ and /o/. Additionally, significant differences were observed when syllables were grouped by consonant place of articulation and manner of articulation. The three platforms performed well for healthy volunteers and maxillectomy patients with their obturator, but the SER for maxillectomy patients without their obturator was high, rendering the platforms unusable. System improvement is needed to increase accuracy for maxillectomy patients.

Speaker-Attributed Training for Multi-Speaker Speech Recognition Using Multi-Stage Encoders and Attention-Weighted Speaker Embedding

Automatic speech recognition (ASR) aims at understanding naturally spoken human speech to be used as text inputs to machines. In multi-speaker environments, where multiple speakers are talking simultaneously with a large amount of overlap, a significant performance degradation may occur with conventional ASR systems if they are trained by recordings of single talkers. This paper proposes a multi-speaker ASR method that incorporates speaker embedding information as an additional input. The embedding information for each of the speakers in the training set was extracted as numeric vectors, and all of the embedding vectors were stacked to construct a total speaker profile matrix. The speaker profile matrix from the training dataset enables finding embedding vectors that are close to the speakers of the input recordings in the test conditions, and it helps to recognize the individual speakers’ voices mixed in the input. Furthermore, the proposed method efficiently reuses the decoder from the existing speaker-independent ASR model, eliminating the need for retraining the entire system. Various speaker embedding methods such as i-vector, d-vector, and x-vector were adopted, and the experimental results show 0.33% and 0.95% absolute (3.9% and 11.5% relative) improvements without and with the speaker profile in the word error rate (WER).

Automatic Speech Recognition of Conversational Speech in Individuals With Disordered Speech.

This study examines the effectiveness of automatic speech recognition (ASR) for individuals with speech disorders, addressing the gap in performance between read and conversational ASR. We analyze the factors influencing this disparity and the effect of speech mode-specific training on ASR accuracy. Recordings of read and conversational speech from 27 individuals with various speech disorders were analyzed using both (a) one speaker-independent ASR system trained and optimized for typical speech and (b) multiple ASR models that were personalized to the speech of the participants with disordered speech. Word error rates were calculated for each speech model, read versus conversational, and subject. Linear mixed-effects models were used to assess the impact of speech mode and disorder severity on ASR accuracy. We investigated nine variables, classified as technical, linguistic, or speech impairment factors, for their potential influence on the performance gap. We found a significant performance gap between read and conversational speech in both personalized and unadapted ASR models. Speech impairment severity notably impacted recognition accuracy in unadapted models for both speech modes and in personalized models for read speech. Linguistic attributes of utterances were the most influential on accuracy, though atypical speech characteristics also played a role. Including conversational speech samples in model training notably improved recognition accuracy. We observed a significant performance gap in ASR accuracy between read and conversational speech for individuals with speech disorders. This gap was largely due to the linguistic complexity and unique characteristics of speech disorders in conversational speech. Training personalized ASR models using conversational speech significantly improved recognition accuracy, demonstrating the importance of domain-specific training and highlighting the need for further research into ASR systems capable of handling disordered conversational speech effectively.

Electrolaryngeal speech enhancement based on a two stage framework with bottleneck feature refinement and voice conversion

An electrolarynx (EL) is a medical device that generates speech for people who lost their biological larynx. However, EL speech signals are unnatural and unintelligible due to the monotonous pitch and the mechanical excitation of the EL device. This paper proposes an end-to-end voice conversion method to enhance EL speech. We adopt a speaker-independent automatic speech recognition model to extract bottleneck features as the intermediate phonetic features for enhancement. Our system includes two stages: the bottleneck feature vectors of the EL speech are mapped by a parallel non-autoregressive model to the corresponding feature vectors of the normal speech in stage one. Then another voice conversion model maps normal speech’s bottleneck feature vectors directly to normal speech’s Mel-spectrogram in stage two, followed by a MelGAN-based vocoder to convert the Mel-spectrogram into waveform. In addition, we incorporate data augmentation and transfer learning to improve conversion performance. Experimental results show that the proposed method outperforms our baseline methods and performs well in terms of naturalness and intelligibility. The audio samples are available online.22https://haydencaffrey.github.io/el/index.html.

Self-segmentation of pass-phrase utterances for deep feature learning in text-dependent speaker verification

In this paper, we propose a novel method to segment and label pass-phrase utterances for training deep neural network (DNN) bottleneck (BN) features for text-dependent speaker verification (TD-SV). Specifically, gender-dependent hidden Markov models (HMMs) for monophones are first trained using the pass-phrase utterances that are disjoint from evaluation. Next, the trained HMMs are speaker-adapted and then used for segmenting and labeling these training utterances at the phone level. The resulted labeled data is subsequently used for training DNN models to discriminate gender-dependent phones for the purpose of extracting phone-discriminant BN features. This is in contrast to conventional approaches that apply a general-purpose, speaker-independent automatic speech recognition (ASR) system for generating segmentation and labels. The proposed method eliminates the need for a separate ASR system, which can additionally have the disadvantage of mismatch with the pass-phrase utterances in terms languages, dialects, domains, acoustic conditions and so on. Experiments are conducted on the RedDots challenge 2016 database of TD-SV using short utterances with Gaussian mixture model-universal background model and i-vector techniques. Experimental results demonstrate that the proposed method yields lower error rates in TD-SV when compared to a set of existing methods. A thorough ablation study further confirms the effectiveness of the method. Fusion in both score and feature levels also shows the complementary nature of the proposed features.

Estimation of ASR Parameterization for Interactive System

In this study, the authors explore the integration of speaker-independent automatic Amazigh speech recognition technology into interactive applications to extract data remotely from a distance database. Based on the combined interactive voice response (IVR) and automatic speech recognition (ASR) technologies, the authors built an interactive speech system to allow users to interact with the interactive system through voice commands. The hidden Markov models (HMMs), Gaussian mixture models (GMMs), and Mel frequency spectral coefficients (MFCCs) are used to develop a speech system based on the ten first Amazigh digits and six Amazigh words. The best-obtained performance is 89.64% by using 3 HMMs and 16 GMMs.

Creating speaker independent ASR system through prosody modification based data augmentation

In this paper, the effect of prosody-modification-based data augmentation is explored in the context of automatic speech recognition (ASR). The primary motive is to develop ASR systems that are less affected by speaker-dependent acoustic variations. Two factors contributing towards inter-speaker variability that are focused on in this paper are pitch and speaking-rate variations. In order to simulate such an ASR task, we have trained an ASR system on adults’ speech and tested it using speech data from adult as well as child speakers. Compared to adults’ speech test case, the recognition rates are noted to be extremely degraded when the test speech is from child speakers. The observed degradation is basically due to large differences in pitch and speaking-rate between adults’ and children’s speech. To overcome this problem, pitch and speaking-rate of the training speech are modified to create new versions of the data. The original and the modified versions are then pooled together in order to capture greater acoustic variability. The ASR system trained on augmented data is noted to be more robust towards speaker-dependent variations. Relative improvements of 11.5% and 27.0% over the baseline are obtained on decoding adults’ and children’s speech test sets, respectively.

Developing speaker independent ASR system using limited data through prosody modification based on fuzzy classification of spectral bins

The primary motive of this study is to develop an automatic speech recognition (ASR) system using limited amount of speech data such that it is least affected by speaker-dependent acoustic variations. The two factors contributing towards inter-speaker variability that are focused upon in this work are pitch and speaking-rate variations. In order to simulate such a limited data scenario, an ASR system is trained on adults' speech and tested using speech data from adult as well as child speakers. Compared to adults' speech test case, the recognition rates are noted to be extremely degraded when the test speech is from child speakers. The observed degradation is due to large differences in pitch and speaking-rate between adults' and children's speech along with other factors leading to inter-speaker acoustic variations. To overcome the mismatch in pitch and speaking-rate, two different approaches are proposed in this paper. In the first approach, the pitch and speaking-rate of children's speech test set are explicitly modified using a recently proposed prosody modification technique that exploits fuzzy classification of spectral bins. In the second approach, pitch and speaking-rate of the training data are modified to create newer versions of the data. In order to capture greater acoustic variability, the original and the modified versions are then pooled together. The ASR system trained on augmented data is noted to be more robust towards pitch and speaking-rate variations. Consequently, relative improvements of 17% and 31% over the baseline are obtained on decoding adults' and children's speech test sets, respectively.

Articulatory and bottleneck features for speaker-independent ASR of dysarthric speech

The rapid population aging has stimulated the development of assistive devices that provide personalized medical support to the needies suffering from various etiologies. One prominent clinical application is a computer-assisted speech training system which enables personalized speech therapy to patients impaired by communicative disorders in the patient’s home environment. Such a system relies on the robust automatic speech recognition (ASR) technology to be able to provide accurate articulation feedback. With the long-term aim of developing off-the-shelf ASR systems that can be incorporated in clinical context without prior speaker information, we compare the ASR performance of speaker-independent bottleneck and articulatory features on dysarthric speech used in conjunction with dedicated neural network-based acoustic models that have been shown to be robust against spectrotemporal deviations. We report ASR performance of these systems on two dysarthric speech datasets of different characteristics to quantify the achieved performance gains. Despite the remaining performance gap between the dysarthric and normal speech, significant improvements have been reported on both datasets using speaker-independent ASR architectures.

An automatic speech recognition system for spontaneous Punjabi speech corpus

Automatic speech recognition is the central part of the wheel towards the natural person-to-machine interaction technique. Due to the high disparity of speaking styles, speech recognition surely demands composite methods to constitute this irregularity. A speech recognition method can work in numerous distinct states such as speaker dependent/independent speech, isolated/continuous/spontaneous speech recognition, for less to very large vocabulary. The Punjabi language is being spoken by concerning 104 million peoples in India, Pakistan and other countries with Punjabi migrants. The Punjabi language is written in Gurmukhi writing in Indian Punjab, while in Shahmukhi writing in Pakistani Punjab. In the paper, the objective is to build the speaker independent automatic spontaneous speech recognition system for the Punjabi language. The system is also capable to recognize the spontaneous Punjabi live speech. So far, no work has to be achieved in the area of spontaneous speech recognition system for the Punjabi language. The user interfaces for Punjabi live speech system is created by using the java programming. Till now, automatic speech system is trained with 6012 Punjabi words and 1433 Punjabi sentences. The performance measured in terms of recognition accuracy which is 93.79% for Punjabi words and 90.8% for Punjabi sentences.

A Regression Approach to Single-Channel Speech Separation Via High-Resolution Deep Neural Networks

We propose a novel data-driven approach to single-channel speech separation based on deep neural networks (DNNs) to directly model the highly nonlinear relationship between speech features of a mixed signal containing a target speaker and other interfering speakers. We focus our discussion on a semisupervised mode to separate speech of the target speaker from an unknown interfering speaker, which is more flexible than the conventional supervised mode with known information of both the target and interfering speakers. Two key issues are investigated. First, we propose a DNN architecture with dual outputs of the features of both the target and interfering speakers, which is shown to achieve a better generalization capability than that with output features of only the target speaker. Second, we propose using a set of multiple DNNs, each intending to be signal-noise-dependent (SND), to cope with the difficulty that one single general DNN could not well accommodate all the speaker mixing variabilities at different signal-to-noise ratio (SNR) levels. Experimental results on the speech separation challenge (SSC) data demonstrate that our proposed framework achieves better separation results than other conventional approaches in a supervised or semisupervised mode. SND-DNNs could also yield significant performance improvements over a general DNN for speech separation in low SNR cases. Furthermore, for automatic speech recognition (ASR) following speech separation, this purely front-end processing with a single set of speaker-independent ASR acoustic models, achieves a relative word error rate (WER) reduction of 11.6% over a state-of-the-art separation and recognition system where a complicated joint back-end decoding framework with multiple sets of speaker-dependent ASR acoustic models needs to be implemented. When speaker-adaptive ASR acoustic models for the target speakers are adopted for the enhanced signals, another 12.1% WER reduction over our best speaker-independent ASR system is achieved.

IMPLEMENTATION OF HIERARCHICAL PHONEME CLASSIFICATION APPROACH ON LTDIGITS CORPORA

Better discrimination of phonemic units still remains one of the most important problems in automatic speech recognition. Direct phoneme recognition in speaker independent automatic speech recognition systems is unable to provide good enough recognition results. There is made an assumption that better results could be achieved through the recognition of phoneme groups using group characteristic features: voiced/unvoiced, vowel/consonant, etc. This paper describes and statistically motivates features and rules for the detection of phoneme groups using phonetically labeled data. Algorithms for recognition of stop and fricative consonants are presented. Experimental research confirmed the advantages of the hierarchical classification of phonemes. Combination of knowledge and rules for detection of acoustic events with the classical statistical classification methods produced an overall 3% improvement of phoneme recognition accuracy and a 52-55% reduction of time taken by classification.

Articulation based admissible wavelet packet feature based on human cochlear frequency response for TIMIT speech recognition

To deal with non-stationary and quasi-stationary signals, wavelet transform has been used as an effective tool for the time-frequency analysis. In the recent years, wavelet transform has been used extensively for feature extraction in noisy speech recognition. These filters have the benefit of having frequency bands spacing similar to the auditory Equivalent Rectangular Bandwidth (ERB) scale. Central frequencies of ERB are equally distributed with the frequency response of the human cochlea.This paper deals with the speaker-independent Automatic Speech Recognition (ASR) system for continuous speech. This Hidden Markov Model (HMM) based ASR system was developed for English using recordings of four regions taken from TIMIT database. A new set of features were derived using wavelet packet transform’s multi-resolution capabilities and having an advantage of ERB filter based on the human cochlea. New set of wavelet features have shown significant improvements in the noisy environment, especially at low SNR values.

Artificial neural networks as speech recognisers for dysarthric speech: Identifying the best-performing set of MFCC parameters and studying a speaker-independent approach

Dysarthria is a neurological impairment of controlling the motor speech articulators that compromises the speech signal. Automatic Speech Recognition (ASR) can be very helpful for speakers with dysarthria because the disabled persons are often physically incapacitated. Mel-Frequency Cepstral Coefficients (MFCCs) have been proven to be an appropriate representation of dysarthric speech, but the question of which MFCC-based feature set represents dysarthric acoustic features most effectively has not been answered. Moreover, most of the current dysarthric speech recognisers are either speaker-dependent (SD) or speaker-adaptive (SA), and they perform poorly in terms of generalisability as a speaker-independent (SI) model. First, by comparing the results of 28 dysarthric SD speech recognisers, this study identifies the best-performing set of MFCC parameters, which can represent dysarthric acoustic features to be used in Artificial Neural Network (ANN)-based ASR. Next, this paper studies the application of ANNs as a fixed-length isolated-word SI ASR for individuals who suffer from dysarthria. The results show that the speech recognisers trained by the conventional 12 coefficients MFCC features without the use of delta and acceleration features provided the best accuracy, and the proposed SI ASR recognised the speech of the unforeseen dysarthric evaluation subjects with word recognition rate of 68.38%.

Speaker-independent ASR for Modern Standard Arabic: effect of regional accents

This paper deals with speaker-independent Automatic Speech Recognition (ASR) system for continuous speech. This ASR system has been developed for Modern Standard Arabic (MSA) using recordings of six regions taken from ALGerian Arabic Speech Database (ALGASD), and has been designed by using Hidden Markov Models.

Studies on inter-speaker variability in speech and its application in automatic speech recognition

In this paper, we give an overview of the problem of inter-speaker variability and its study in many diverse areas of speech signal processing. We first give an overview of vowel-normalization studies that minimize variations in the acoustic representation of vowel realizations by different speakers. We then describe the universal-warping approach to speaker normalization which unifies many of the vowel normalization approaches and also shows the relation between speech production, perception and auditory processing. We then address the problem of inter-speaker variability in automatic speech recognition (ASR) and describe techniques that are used to reduce these effects and thereby improve the performance of speaker-independent ASR systems.

Speech-Audio Interface for Medical Information Management in Field Environments

The outcome of emergency medical care often depends on how quickly and accurately field medical personnel can access patient background information, document their assessment and treatment of the patient, and communicate this information to those who will next receive the patient. These tasks pose a difficult challenge to field medical personnel, who must devote their eyes and hands to patients rather than to documentation. A prototype system was designed through User-centered Systems Engineering to overcome physical and cognitive constraints on mobile field medical providers' information management capabilities. This solution involved electronic patient records that could be updated through spoken inputs to a speaker-independent automatic speech recognition system and that could be accessed through spoken requests for synthesized speech outputs of their contents. To ensure rapid and accurate interpretation of spoken inputs, the system incorporated a grammar and restricted vocabulary modeled after the grammar and vocabulary spontaneously used by medics to describe medical incidents and the vocabulary on paper patient records. When tested by males and females with a variety of accents in a quiet environment, the system typically recognized more than 90% of the inputs on the first try. With background noises characteristic of field medical situations, recognition percentages were typically in the 70s and 80s. User acceptance was high, because the system allowed eyes-free, hands-free operation with speech that seemed natural, provided multiple means for remedying inputs that were not correctly interpreted, and digitally recorded the user's description of the incident.

A two-pass hybrid system using a low dimensional auditory model for speaker-independent isolated-word recognition

Including speech knowledge in automatic speech recognition (ASR) systems is a good way to improve the performance of recognizers. In this paper, we propose the orion system which deals with speaker-independent ASR for isolated-words. orion is a two-pass hybrid system which uses several types of knowledge. This knowledge applies to psychoacoustics, physiology and phonetics. During the first pass, an auditory model, the perceptually-based linear prediction analysis (PLP), combines static and dynamic features to provide a set of parameters to the dynamic programming algorithm. After this stage 98% recognition accuracy was obtained for a digit vocabulary and 12 templates per word. In the case of a confusable vocabulary ( E-SET), the introduction of phonetic knowledge in the second pass decreases the error rate by more than 60% (compared to the results of the first pass).

Perceptual linear predictive (PLP) analysis of speech

A new technique for the analysis of speech, the perceptual linear predictive (PLP) technique, is presented and examined. This technique uses three concepts from the psychophysics of hearing to derive an estimate of the auditory spectrum: (1) the critical-band spectral resolution, (2) the equal-loudness curve, and (3) the intensity-loudness power law. The auditory spectrum is then approximated by an autoregressive all-pole model. A 5th-order all-pole model is effective in suppressing speaker-dependent details of the auditory spectrum. In comparison with conventional linear predictive (LP) analysis, PLP analysis is more consistent with human hearing. The effective second formant F2' and the 3.5-Bark spectral-peak integration theories of vowel perception are well accounted for. PLP analysis is computationally efficient and yields a low-dimensional representation of speech. These properties are found to be useful in speaker-independent automatic-speech recognition.

Comparison of standard ASR front ends and auditory models in neural net-based automatic speech recognition

Recent work [Renals and Terry, European Conference on Speech Technology, Edinburgh, United Kingdom (1987)] reported on a small multispeaker isolated digit automatic speech recognition (ASR) experiment using a backward propagation neural net system with a synchrony auditory model as its front end. In spite of fairly crude temporal normalization, the system was capable of better than 80% ASR accuracy. The question remains if the reported performance is due to (a) the paticular front end, (b) the particular neural net-based classification, (c) the particular time-normalization scheme, or (d) the combination of all factors. The goal here is to isolate these factors. In the present contribution, different front ends in the neural net-based ASR are systematically evaluated. Several standard ASR front ends are compared with the synchrony auditory model and the perceptually based linear predictive auditory model front ends in both the speaker-dependent and the speaker-independent ASR. The speed of learning and the ASR accuracy of the compared recognizers are reported and discussed.