Improving Recognition of Dysarthric Speech Using Severity Based Tempo Adaptation

Bhat, Chitralekha; Vachhani, Bhavik; Kopparapu, Sunil Kumar

doi:10.1007/978-3-319-43958-7_44

Cited by 8 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to modify healthy control speech to emulate dysarthric speech characteristics, we need to first understand the dysarthric speech itself. In our earlier work [23], we modified the tempo of dysarthric speech based on severity to improve the ASR recognition. It was observed that the sonorant regions of dysarthric speech are of longer durations as compared to that of healthy speech.…”

Section: Methodology 21 Phoneme Duration Analysismentioning

confidence: 99%

Data Augmentation Using Healthy Speech for Dysarthric Speech Recognition

Vachhani¹,

Bhat²,

Kopparapu³

2018

Interspeech 2018

Self Cite

View full text Add to dashboard Cite

Dysarthria refers to a speech disorder caused by trauma to the brain areas concerned with motor aspects of speech giving rise to effortful, slow, slurred or prosodically abnormal speech. Traditional Automatic Speech Recognizers (ASR) perform poorly on dysarthric speech recognition tasks, owing mostly to insufficient dysarthric speech data. Speaker related challenges complicates data collection process for dysarthric speech. In this paper, we explore data augmentation using temporal and speed modifications to healthy speech to simulate dysarthric speech. DNN-HMM based Automatic Speech Recognition (ASR) and Random Forest based classification were used for evaluation of the proposed method. Dysarthric speech, generated synthetically, is classified for severity level using a Random Forest classifier that is trained on actual dysarthric speech. ASR trained on healthy speech, augmented with simulated dysarthric speech is evaluated for dysarthric speech recognition. All evaluations were carried out using Universal Access dysarthric speech corpus. An absolute improvement of 4.24% and 2% WAS achieved using tempo based and speed based data augmentation respectively as compared to ASR performance using healthy speech alone for training.

show abstract

Section: Methodology 21 Phoneme Duration Analysismentioning

confidence: 99%

Data Augmentation Using Healthy Speech for Dysarthric Speech Recognition

Vachhani¹,

Bhat²,

Kopparapu³

2018

Interspeech 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…We propose a method to improve the recognition of dysarthric speech using enhanced speech features that have been extracted using a Deep Autoencoder (DAE). Additionally, we extend our earlier work [21], wherein we transform the dysarthric speech in the temporal domain using severity-based tempo adaptation (TA) and use the tempo adapted dysarthric speech prior to feature enhancement using a DAE. We analyse the contribution of the individual techniques towards improvement in speech recognition as well as tempo adaptation and DAE-based feature enhancement in tandem.…”

Section: Introductionmentioning

confidence: 88%

“…Tempo adaptation parameters as shown in Table 1, were empirically determined for different severity levels in the UA speech corpus as described in [21]. We use the Kaldi [25] toolkit-based deep autoencoder for our experiments.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Autoencoder Based Speech Features for Improved Dysarthric Speech Recognition

et al. 2017

Self Cite

View full text Add to dashboard Cite

Dysarthria is a motor speech disorder, resulting in mumbled, slurred or slow speech that is generally difficult to understand by both humans and machines. Traditional Automatic Speech Recognizers (ASR) perform poorly on dysarthric speech recognition tasks. In this paper, we propose the use of deep autoencoders to enhance the Mel Frequency Cepstral Coefficients (MFCC) based features in order to improve dysarthric speech recognition. Speech from healthy control speakers is used to train an autoencoder which is in turn used to obtain improved feature representation for dysarthric speech. Additionally, we analyze the use of severity based tempo adaptation followed by autoencoder based speech feature enhancement. All evaluations were carried out on Universal Access dysarthric speech corpus. An overall absolute improvement of 16% was achieved using tempo adaptation followed by autoencoder based speech front end representation for DNN-HMM based dysarthric speech recognition.

show abstract

“…Acoustic space modification carried out through temporal and frequency morphing improved automatic dysarthric speech recognition as well as subjective evaluation in [3]. It can be seen that temporal adaptation based on dysarthria severity level improved the ASR performamce for dysarthric speech recognition at each severity level [4]. A Convolutive Bottleneck Network (CBN) was used for dysarthric speech feature extraction wherein the pooling operations of the CBN resulted in features that were more robust towards the small local fluctuations in dysarthric speech and outperformed the traditional MFCC feature based recognition [5].…”

Section: Introductionmentioning

confidence: 91%

Dysarthric Speech Recognition Using Time-delay Neural Network Based Denoising Autoencoder

et al. 2018

Self Cite

View full text Add to dashboard Cite

Dysarthria is a manisfestation of the disruption in the neuromuscular physiology resulting in uneven, slow, slurred, harsh or quiet speech. Dysarthric speech poses serious challenges to automatic speech recognition, considering this speech is difficult to decipher for both humans and machines. The objective of this work is to enhance dysarthric speech features to match that of healthy control speech. We use a Time-Delay Neural Network based Denoising Autoencoder (TDNN-DAE) to enhance the dysarthric speech features. The dysarthric speech thus enhanced is recognized using a DNN-HMM based Automatic Speech Recognition (ASR) engine. This methodology was evaluated for speaker-independent (SI) and speaker-adapted (SA) systems. Absolute improvements of 13% and 3% was observed in the ASR performance for SI and SA systems respectively as compared with unenhanced dysarthric speech recognition.

show abstract

Improving Recognition of Dysarthric Speech Using Severity Based Tempo Adaptation

Cited by 8 publications

References 13 publications

Data Augmentation Using Healthy Speech for Dysarthric Speech Recognition

Data Augmentation Using Healthy Speech for Dysarthric Speech Recognition

Deep Autoencoder Based Speech Features for Improved Dysarthric Speech Recognition

Dysarthric Speech Recognition Using Time-delay Neural Network Based Denoising Autoencoder

Contact Info

Product

Resources

About