Sequence labeling to detect stuttering events in read speech

Alharbi, Sadeen; Hasan, Madina; Simons, Anthony J. H.; Brumfitt, Shelagh; Green, Phil

doi:10.1016/j.csl.2019.101052

Cited by 18 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sadeen et. al [20] studied two approaches to do sequence labeling for stuttered speech, Conditional Random Fields (CRF) and Bi-directional Long Short Term Memory Networks (BI-LSTM). The study showed an improvement in F1 Score by 33.6% for the BiL-STM model relative to CRF baseline, using ngram based features.…”

Section: Related Workmentioning

confidence: 99%

Enhancing ASR for Stuttered Speech with Limited Data Using Detect and Pass

Shonibare¹,

Tong²,

Ravichandran³

2022

Preprint

View full text Add to dashboard Cite

It is estimated that around 70 million people worldwide are affected by a speech disorder called stuttering [1]. With recent advances in Automatic Speech Recognition (ASR), voice assistants are increasingly useful in our everyday lives. Many technologies in education, retail, telecommunication and healthcare can now be operated through voice. Unfortunately, these benefits are not accessible for People Who Stutter (PWS). We propose a simple but effective method called 'Detect and Pass' to make modern ASR systems accessible for People Who Stutter in a limited data setting. The algorithm uses a context aware classifier trained on a limited amount of data, to detect acoustic frames that contain stutter. To improve robustness on stuttered speech, this extra information is passed on to the ASR model to be utilized during inference. Our experiments show a reduction of 12.18% to 71.24% in Word Error Rate (WER) across various state of the art ASR systems. Upon varying the threshold of the associated posterior probability of stutter for each stacked frame used in determining low frame rate (LFR) acoustic features, we were able to determine an optimal setting that reduced the WER by 23.93% to 71.67% across different ASR systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Enhancing ASR for Stuttered Speech with Limited Data Using Detect and Pass

Shonibare¹,

Tong²,

Ravichandran³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Following the Bayesian statistical framework, this entails maximization of the a-posteriori probability of the word sequence with disfluencies, given the originally intended word sequence. Since disfluency events affect different phonetic aspects of speech [6] many researchers tried to take advantage of the possible combination of different sources of knowledge on both acoustic and language model sides [7,8,9,10]. The disfluency detection task in all mentioned works is solved by using sequence labelling/tagging approaches which can rely either on a generative approach such as Hidden Markov Model or discriminative log-linear models, such as Maximum Entropy Markov Model or Conditional Random Fields as well as Bidirectional Long Short-Term Memory based networks in combination with an attention mechanism [11].…”

Section: Prior Workmentioning

confidence: 99%

Improved Robustness to Disfluencies in Rnn-Transducer Based Speech Recognition

Mendelev

Raissi

Camporese

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Automatic Speech Recognition (ASR) based on Recurrent Neural Network Transducers (RNN-T) is gaining interest in the speech community. We investigate data selection and preparation choices aiming for improved robustness of RNN-T ASR to speech disfluencies with a focus on partial words. For evaluation we use clean data, data with disfluencies and a separate dataset with speech affected by stuttering. We show that after including a small amount of data with disfluencies in the training set the recognition accuracy on the tests with disfluencies and stuttering improves. Increasing the amount of training data with disfluencies gives additional gains without degradation on the clean data. We also show that replacing partial words with a dedicated token helps to get even better accuracy on utterances with disfluencies and stutter. The evaluation of our best model shows 22.5% and 16.4% relative WER reduction on those two evaluation sets.

show abstract

“…Stuttering identification, an interdisciplinary research problem in which a myriad number of research work (in-terms of acoustic feature extraction and classification methods) are currently going on with a focus on developing automatic tools for its detection and identification. Most of the existing work detect and identify stuttering either by language models [73,74] or by ASR systems [75,76], which first converts the audio signals into its corresponding textual form, and then by the application of language models, detects or identifies stuttering. This section provides in detail the comprehensive review of the various acoustic based feature extraction and machine learning stuttering identification techniques, that have been used in the literature.…”

Section: Statistical Approachesmentioning

confidence: 99%

Machine Learning for Stuttering Identification: Review, Challenges and Future Directions

Sheikh¹,

Sahidullah²,

Hirsch³

et al. 2021

Preprint

View full text Add to dashboard Cite

Stuttering is a speech disorder during which the flow of speech is interrupted by involuntary pauses and repetition of sounds. Stuttering identification is an interesting interdisciplinary domain research problem which involves pathology, psychology, acoustics, and signal processing that makes it hard and complicated to detect. Recent developments in machine and deep learning have dramatically revolutionized speech domain, however minimal attention has been given to stuttering identification. This work fills the gap by trying to bring researchers together from interdisciplinary fields. In this paper, we review comprehensively acoustic features, statistical and deep learning based stuttering/disfluency classification methods. We also present several challenges and possible future directions.

show abstract

Sequence labeling to detect stuttering events in read speech

Cited by 18 publications

References 29 publications

Enhancing ASR for Stuttered Speech with Limited Data Using Detect and Pass

Enhancing ASR for Stuttered Speech with Limited Data Using Detect and Pass

Improved Robustness to Disfluencies in Rnn-Transducer Based Speech Recognition

Machine Learning for Stuttering Identification: Review, Challenges and Future Directions

Contact Info

Product

Resources

About