Context-independent Models Research Articles

Development and analysis of Punjabi ASR system for mobile phones under different acoustic models

Speech technology is widely gaining importance in our daily life. Speech based mobile phone applications are becoming popular in masses due to their usability and ease of access. Speech technology is helping people, with disabilities like blindness and physical abnormalities, to access and control mobile phone applications through voice, without using keypad or touchpad. Punjabi is one of the widely spoken language in various parts of the world. In this paper, an automatic speech recognition (ASR) system for mobile phone applications in Punjabi has been proposed and implemented for four different acoustic models- context independent, context dependent untied, context dependent tied, and context dependent deleted interpolation models. The proposed ASR is evaluated at 4, 16, 32 and 64 GMMs for performance analysis in terms of parameters like accuracy, word error rate and storage space required. It is observed that context dependent untied models outperform others by having better accuracy and lower word error rate, while context independent models require less storage space than others. The choice of fruitful acoustic model depends upon the available storage space as well as desired recognition accuracy. Mobile phones having limited resources may use context independent models, while context dependent untied models can be used to develop ASR system for high end mobile phones.

Combination of context-dependent bidirectional long short-term memory classifiers for robust offline handwriting recognition

The BLSTM classifier has been recently introduced for sequence labeling tasks and provides state-of-the-art performance for handwriting recognition. Its recurrent connections integrate the context at the feature level over a long range. Nevertheless, this context is not explicitly modeled at the label level. Explicit context-modeling strategies have been applied to HMMs with improvement of the recognition rate. In this paper, we study the effect of context modeling on the performance of the BLSTM classifier. The baseline approach, consisting of context-independent character label, is compared with several context-dependent approaches, modeling the left and right contexts. The results show that context-dependent models improve the recognition rate, and demonstrate the ability of the BLSTM classifier to deal with a large number of character models, without clustering. Furthermore, the context-dependent and context-independent models are complementary, and their combination leads to a robust recognition. We tested our approach with promising results on the RIMES database of Latin script documents.

Robust Bayesian estimation for context-based speech enhancement

Abstract Model-based speech enhancement algorithms that employ trained models, such as codebooks, hidden Markov models, Gaussian mixture models, etc., containing representations of speech such as linear predictive coefficients, mel-frequency cepstrum coefficients, etc., have been found to be successful in enhancing noisy speech corrupted by nonstationary noise. However, these models are typically trained on speech data from multiple speakers under controlled acoustic conditions. In this paper, we introduce the notion of context-dependent models that are trained on speech data with one or more aspects of context, such as speaker, acoustic environment, speaking style, etc. In scenarios where the modeled and observed contexts match, context-dependent models can be expected to result in better performance, whereas context-independent models are preferred otherwise. In this paper, we present a Bayesian framework that automatically provides the benefits of both models under varying contexts. As several aspects of the context remain constant over an extended period during usage, a memory-based approach that exploits information from past data is employed. We use a codebook-based speech enhancement technique that employs trained models of speech and noise linear predictive coefficients as an example model-based approach. Using speaker, acoustic environment, and speaking style as aspects of context, we demonstrate the robustness of the proposed framework for different context scenarios, input signal-to-noise ratios, and number of contexts modeled.

Topological invariants as speech features for automatic speech recognition

The article presents topological invariants as speech features for speech recognition systems based on hidden Markov models. A short introduction is provided to the mathematical concept of topological invariants and space symmetries for the speech recognition problem. This involves a basic overview of the relevant auditory characteristic and its modelling in order to identify possible symmetries and invariants. Once the concept is derived, several of its modifications vital for HMM systems such as reduction of dimensions, within–class feature decorrelation and a signal plane rotation are presented and evaluated on a real system. The final system is evaluated and compared to other features using both context–dependent and context–independent models. Tests were accomplished on the professional speech database, where the achieved accuracies reached up to 97.7%, 98.7% and 98.9% for string of digits, application words and isolated digits tests, respectively.

Robust Bayesian estimation for context-based speech enhancement

Model-based speech enhancement algorithms that employ trained models, such as codebooks, hidden Markov models, Gaussian mixture models, etc., containing representations of speech such as linear predictive coefficients, mel-frequency cepstrum coefficients, etc., have been found to be successful in enhancing noisy speech corrupted by nonstationary noise. However, these models are typically trained on speech data from multiple speakers under controlled acoustic conditions. In this paper, we introduce the notion of context-dependent models that are trained on speech data with one or more aspects of context, such as speaker, acoustic environment, speaking style, etc. In scenarios where the modeled and observed contexts match, context-dependent models can be expected to result in better performance, whereas context-independent models are preferred otherwise. In this paper, we present a Bayesian framework that automatically provides the benefits of both models under varying contexts. As several aspects of the context remain constant over an extended period during usage, a memory-based approach that exploits information from past data is employed. We use a codebook-based speech enhancement technique that employs trained models of speech and noise linear predictive coefficients as an example model-based approach. Using speaker, acoustic environment, and speaking style as aspects of context, we demonstrate the robustness of the proposed framework for different context scenarios, input signal-to-noise ratios, and number of contexts modeled.

Improved syllable-based acoustic modeling for continuous Chinese speech recognition

Concerning the changeability of the speech signal caused by co-articulation phenomenon in Chinese speech recognition,a syllable-based acoustic modeling method was proposed.Firstly,context independent syllable-based acoustic models were trained,and the models were initialized by intra-syllable IFs based diphones to solve the problem of training data sparsity.Secondly,the inter-syllable co-articulation effect was captured by incorporating inter-syllable transition models into the recognition system.The experiments conducted on "863-test" dataset show that the relative character error rate is reduced by 12.13%.This proves that syllable-based acoustic model and inter-syllable transition model are effective in solving co-articulation effect.

Development & evaluation of different acoustic models for Malayalam continuous speech recognition

Performance of any continuous speech recognition system is dependent on the accuracy of its acoustic model. Hence, preparation of a robust and accurate acoustic model lead to satisfactory recognition performance for a speech recognizer. In acoustic modeling of phonetic unit, context information is of prime importance as the phonemes are found to vary according to the place of occurrence in a word. In this paper we compare and evaluate the effect of context dependent tied (CD tied) models, context dependent (CD) and context independent (CI) models in the perspective of continuous speech recognition of Malayalam language. The database for the speech recognition system has utterance from 21 speakers including 11 female and 10 males. Our evaluation results show that CD tied models outperforms CI models over 21%.

Combining formant frequency based on variable order LPC coding with acoustic features for TIMIT phone recognition

Combination of multiple acoustic features has great potential to improve Automatic Speech Recognition (ASR) accuracy. Our contribution in this research was to investigate one novel method when using voiced formants' features in combination with standard MFCC features in order to enhance TIMIT phone recognition. These voiced features provide accurate formants frequencies using a Variable Order LPC Coding (VO-LPC) algorithm that was combined with continuity constraints. The overall estimating formants were concatenated with MFCC features when a voiced frame could be detected. For feature-level combination, Heteroscedastic Linear Discriminant Analysis (HLDA) based approach had been used successfully to find an optimal linear combination of successive vectors of a single feature stream. A series of experiments on phone recognition speaker-independent continuous-speech had been carried out using a subset of the large read-speech TIMIT phone corpus. Hidden Markov Model Toolkit (HTK) was also used throughout all carried experiments. Using such feature level combination, optimized mixture splitting and a bigram language model, a detailed analysis on this combination performance was discussed for Context-Independent (CI) and Context-Dependent (CD) Hidden Markov Models (HMM). Experimental results from our proposed procedure showed that phone error rate was successfully decreased by about 3%. At phonetic level group, an increase of 8% and of 10% was observed respectively for vowel and liquid group. These results proved clear phone enhancement regarding existing state of the art.

Modelling Tonal Attraction Between Adjacent Musical Elements

A context-independent model of tonal attraction is presented based on the formal musical property of interval cycles. An interval cycle is the minimum number of additive iterations of an interval that are required for the original pitch classes to be re-stated. Interval cycles are conjectured to give rise to an abstract grouping property, interval cycle proximity, which in turn is responsible for tonal attraction. The model was tested using a probe tone experiment requiring subjects to rate the probe for strength of attraction/resolution with respect to a preceding context chord. The results, displayed as ‘attraction profiles’, agreed with the predictions of the model, and showed that even diatonic chords, such as dominant sevenths, can be heard chromatically. The results are discussed in relation to examples of real music and previous research within the field.

Automatic Speech Recognition for Under-Resourced Languages: Application to Vietnamese Language

This paper presents our work in automatic speech recognition (ASR) in the context of under-resourced languages with application to Vietnamese. Different techniques for bootstrapping acoustic models are presented. First, we present the use of acoustic-phonetic unit distances and the potential of crosslingual acoustic modeling for under-resourced languages. Experimental results on Vietnamese showed that with only a few hours of target language speech data, crosslingual context independent modeling worked better than crosslingual context dependent modeling. However, it was outperformed by the latter one, when more speech data were available. We concluded, therefore, that in both cases, crosslingual systems are better than monolingual baseline systems. The proposal of grapheme-based acoustic modeling, which avoids building a phonetic dictionary, is also investigated in our work. Finally, since the use of sub-word units (morphemes, syllables, characters, etc.) can reduce the high out-of-vocabulary rate and improve the lack of text resources in statistical language modeling for under-resourced languages, we propose several methods to decompose, normalize and combine word and sub-word lattices generated from different ASR systems. The proposed lattice combination scheme results in a relative syllable error rate reduction of 6.6% over the sentence MAP baseline method for a Vietnamese ASR task.

Syllable modeling in continuous speech recognition for Tamil language

In automatic speech recognition, the phone has probably been a dominating sub-word unit for more than one decade. Context Dependent phone or triphone modeling accounts for contextual variations between adjacent phones and state tying addresses modeling of triphones that are not seen during training. Recently, syllable is gaining momentum as a new sub-word unit. Syllable being a larger unit than a phone addresses the severe contextual variations between phones within it. Therefore, it is more stable than a phone and models pronunciation variability in a systematic way. Tamil language has challenging features like agglutination and morpho-phonology. In this paper, attempts have been made to provide solutions to these issues by using the syllable as a sub-word unit in an acoustic model. Initially, a small vocabulary context independent word models and a medium vocabulary context dependent phone models are developed. Subsequently, an algorithm based on prosodic syllable is proposed and two experiments have been conducted. First, syllable based context independent models have been trained and tested. Despite large number of syllables, this system has performed reasonably well compared to context independent word models in terms of word error rate and out of vocabulary words. Subsequently, in the second experiment, syllable information is integrated in conventional triphone modeling wherein cross-syllable triphones are replaced with monophones and the number of context dependent phone models is reduced by 22.76% in untied units. In spite of reduction in the number of models, the accuracy of the proposed system is comparable to that of the baseline triphone system.

Explicit Transition Modelling for Automatic Singing Transcription

We present a system for the automatic transcription of solo human singing into note sequences. Particular attention is devoted to the modelling of the transitions between consecutive notes of the sung passage. Hidden Markov models (HMMs) are used to represent both individual notes and the transitions between them, in order to capture the variability of the estimated pitch within a statistical framework. A database consisting of annotated passages sung by 26 soprano subjects is compiled and used for the development of the system. Explicit transition models are introduced in order to better identify note boundaries that are otherwise poorly modelled. Context-independent transition models are evaluated first, followed by context-dependent transition models which are shown to improve both the note recognition accuracy and the transition region time-alignment. The final system is found to be able to transcribe sung passages with an overall accuracy of 86.7%.

Real-Time Continuous Phoneme Recognition System Using Class-Dependent Tied-Mixture HMM With HBT Structure for Speech-Driven Lip-Sync

This work describes a real-time lip-sync method using which an avatar's lip shape is synchronized with the corresponding speech signal. Phoneme recognition is generally regarded as an important task in the operation of a real-time lip-sync system. In this work, the use of the Head-Body-Tail (HBT) model is proposed for the purpose of more efficiently recognizing phonemes which are variously uttered due to co-articulation effects. The HBT model effectively deals with the transition parts of context-dependent models for small-sized vocabulary tasks. These models provide better recognition performance than general context-dependent or context-independent models for the task of digit or vowel recognition. Moreover, each phoneme is categorized into one among four classes and the class-dependent codebook is generated to further improve the performance. Additionally, for the clear representation of the context dependency information in the transient parts, some Gaussians are excluded from class-dependent codebook. The proposed method leads to a lip-sync system that performs at a level that is similar to previous designs based on HBT and continuous hidden Markov models (CHMMs). However, our method reduces the number of model parameters by one-third and enables real-time operation.

A speech and character combined recognition engine for mobile devices

PurposeThe purpose of this paper is to describe a speech and character combined recognition engine (SCCRE) developed for working on personal digital assistants (PDAs) or on mobile devices. Also, the architecture of a distributed recognition system for providing a more convenient user interface is discussed.Design/methodology/approachIn SCCRE, feature extraction for speech and for character is carried out separately, but the recognition is performed in an engine. The client recognition engine essentially employs a continuous hidden Markov model (CHMM) structure and this CHMM structure consists of variable parameter topology in order to minimize the number of model parameters and to reduce recognition time. This model also adopts the proposed successive state and mixture splitting (SSMS) method for generating context independent model. SSMS optimizes the number of mixtures through splitting in mixture domain and the number of states through splitting in time domain.FindingsThe recognition results show that the developed engine can reduce the total number of Gaussian up to 40 per cent compared with the fixed parameter models at the same recognition performance when applied to speech recognition for mobile devices. It shows that SSMS can reduce the size of memory for models to 65 per cent and that for processing to 82 per cent. Moreover, the recognition time decreases 17 per cent with the SMS model while maintaining the recognition rate.Originality/valueThe proposed system will be very useful for many on‐line multimodal interfaces such as PDAs and mobile applications.

Estimation of DNA Sequence Context-dependent Mutation Rates Using Primate Genomic Sequences: Application to Estimation of Selection Bias in Protein (Human TP53) Evolution

AbstractUnderstanding of the mechanism of DNA mutation process is critical for studying the functional consequences of genetic variation in clinical medicine (eg. drug response) as well as other complex traits (eg. gene expression and cause of common diseases). During the last several decades, many probabilistic models of DNA nucleotide substitution have been proposed for studying this process. A common feature of these mutation models is that they assume the nucleotides evolve independently at each site. In other words, they are sequence context-independent models. However, based on various biochemical studies, it is now recognized that the DNA mutation process resulting in substitutions in both coding and non-coding regions may depend on sequence context. We proposed here a more realistic sequence context-dependent mutation model, which could contribute to the better understanding of the DNA mutation spectrum in genomes.We also showed its application to protein evolution by separating the mutation bias and selection bias in amino acid substitutions.

An algebra for multicriteria spatial modeling

The objective of this paper is to propose a new algebra, called decision map algebra (DMA), especially devoted to multicriteria spatial modeling. DMA is a generic and context-independent modeling language inspired from map algebra and other similar languages. It supports the conventional geometric objects and cartographic maps and adds a rich set of new operands representing different spatial multicriteria concepts. Moreover, DMA supports different conventional map algebra-like operations and adds several new ones corresponding to different spatial multicriteria evaluation functions. The proposed algebra is the first step towards the development of a generic spatial multicriteria modeling language.

Design and evaluation of acoustic and language models for large scale telephone services

This paper describes the specification, design and development phases of two widely used telephone services based on automatic speech recognition. The effort spent for evaluating and tuning these services will be discussed in detail. In developing the first service, mainly based on the recognition of “alphanumeric” sequences, a significant part of the work consisted in refining the acoustic models. To increase recognition accuracy we adopted algorithms and methods consolidated in the past over broadcast news transcription tasks. A significant result shows that the use of task specific context dependent phone models reduces the word error rate by about 40% relative to using context independent phone models. Note that the latter result was achieved over a small vocabulary task, significantly different from those generally used in broadcast news transcription. We also investigated both unsupervised and supervised training procedures. Moreover, we studied a novel partly supervised technique that allows us to select in some “optimal” way the speech material to manually transcribe and use for acoustic model training. A significant result shows that the proposed procedure gives performance close to that obtained with a completely supervised training method. In the second service, mainly based on phrase spotting, a wide effort was devoted to language model refinement. In particular, several types of rejection networks were studied to detect out of vocabulary words for the given task; a major result demonstrates that using rejection networks based on a class trigram language model reduces the word error rate from 36.7% to 11.1% with respect to using a phone loop network. For the latter service, the benefits and related costs brought by regular grammars, stochastic language models and mixed language models will be also reported and discussed. Finally, notice that most of experiments described in this paper were carried out on field databases collected through the developed services.

On the efficiency of classical RASTA filtering for continuous speech recognition: Keeping the balance between acoustic pre-processing and acoustic modelling

The efficiency of classical RASTA filtering for channel normalisation was investigated for continuous speech recognition based on context-independent and context-dependent hidden Markov models. For a medium and a large vocabulary continuous speech recognition task, recognition performance was established for classical RASTA filtering and compared to using no channel normalisation, cepstrum mean normalisation, and phase-corrected RASTA. Phase-corrected RASTA is a technique that consists of classical RASTA filtering followed by a phase correction operation. In this manner, channel bias is as effectively removed as with classical RASTA. However, for phase-corrected RASTA, amplitude drift towards zero in stationary signal portions is diminished compared to classical RASTA. The results show that application of classical RASTA filtering resulted in decreased recognition performance when compared to using no channel normalisation for all conditions studied, although the decrease appeared to be smaller for context-dependent models than for context-independent models. However, for all conditions, recognition performance was significantly and substantially improved when phase-corrected RASTA was used and reached the same performance level as obtained for cepstrum mean normalisation in some cases. It is concluded that classical RASTA filtering can only be effective for channel robustness, if the impact of the amplitude drift towards zero can be kept as limited as possible.

TONE RECOGNITION OF CONTINUOUS THAI SPEECH UNDER TONAL ASSIMILATION AND DECLINATION EFFECTS USING HALF-TONE MODEL

This paper presents a method for continuous Thai tone recognition. One of the main problems in tone recognition is that several interacting factors affect F0realization of tones. In this paper, we focus on the tonal assimilation and declination effects. These effects are compensated by the tone information of neighboring syllables, the F0downdrift and the context-dependent tone model. However, the context-dependent tone model is too large and its training time is very long. To overcome these problems, we propose a novel model called the half-tone model. The experiments, which compare all tone features and all tone models, were simulated by feedforward neural networks. The results show that the proposed tone features increase the recognition rates and the half-tone model outperforms conventional tone models, i.e. context-independent and context-dependent tone models, in terms of recognition rate and speed. The best results are 94.77% and 93.82% for the inside test and outside test, respectively.

Feature-based speech recognizer having probabilistic linguistic processor providing word matching based on the entire space of feature vectors

A feature-based speech recognizer having a probabilistic linguistic processor provides word matching based on the entire space of feature vectors. In this manner, the errors and inaccuracies associated with the heretofore known feature-based speech recognizers, which provided word matching on less than the entire space of feature vectors, are overcome, thereby resulting in improved-accuracy speech recognition. The word matching may be on feature vectors computed either from segments or from landmarks or from both segments and landmarks. For word matching on segment-based feature vectors, acoustic likelihoods may be normalized by extra-acoustic likelihoods defined by at least one extra-acoustic ("not" or "anti") model. Context-dependent and context-independent acoustic models may be employed.