Loading…
A comparison of model validation techniques for audio-visual speech recognition
This paper implements and compares the performance of a number of techniques proposed for improving the accuracy of Automatic Speech Recognition (ASR) systems. As ASR that uses only speech can be contaminated by environmental noise, in some applications it may improve performance to employ Audio-Vis...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Default Conference proceeding |
| Published: |
2017
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/27016 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1822178488324456448 |
|---|---|
| author | Thum W. Seong M.Z. Ibrahim Nurul W. Arshad David Mulvaney |
| author_facet | Thum W. Seong M.Z. Ibrahim Nurul W. Arshad David Mulvaney |
| author_sort | Thum W. Seong (7208510) |
| collection | Figshare |
| description | This paper implements and compares the performance of a number of techniques proposed for improving the accuracy of Automatic Speech Recognition (ASR) systems. As ASR that uses only speech can be contaminated by environmental noise, in some applications it may improve performance to employ Audio-Visual Speech Recognition (AVSR), in which recognition uses both audio information and mouth movements obtained from a video recording of the speaker’s face region. In this paper, model validation techniques, namely the holdout method, leave-one-out cross validation and bootstrap validation, are implemented to validate the performance of an AVSR system as well as to provide a comparison of the performance of the validation techniques themselves. A new speech data corpus is used, namely the Loughborough University Audio-Visual (LUNA-V) dataset that contains 10 speakers with five sets of samples uttered by each speaker. The database is divided into training and testing sets and processed in manners suitable for the validation techniques under investigation. The performance is evaluated using a range of different signal-to-noise ratio values using a variety of noise types obtained from the NOISEX-92 dataset. |
| format | Default Conference proceeding |
| id | rr-article-9552131 |
| institution | Loughborough University |
| publishDate | 2017 |
| record_format | Figshare |
| spelling | rr-article-95521312017-01-01T00:00:00Z A comparison of model validation techniques for audio-visual speech recognition Thum W. Seong (7208510) M.Z. Ibrahim (7204967) Nurul W. Arshad (7208513) David Mulvaney (1252071) Mechanical engineering not elsewhere classified Audio-visual speech recognition Hidden markov models HTK toolkit Holdout validation Leave-one-out cross validation Bootstrap validation Mechanical Engineering not elsewhere classified This paper implements and compares the performance of a number of techniques proposed for improving the accuracy of Automatic Speech Recognition (ASR) systems. As ASR that uses only speech can be contaminated by environmental noise, in some applications it may improve performance to employ Audio-Visual Speech Recognition (AVSR), in which recognition uses both audio information and mouth movements obtained from a video recording of the speaker’s face region. In this paper, model validation techniques, namely the holdout method, leave-one-out cross validation and bootstrap validation, are implemented to validate the performance of an AVSR system as well as to provide a comparison of the performance of the validation techniques themselves. A new speech data corpus is used, namely the Loughborough University Audio-Visual (LUNA-V) dataset that contains 10 speakers with five sets of samples uttered by each speaker. The database is divided into training and testing sets and processed in manners suitable for the validation techniques under investigation. The performance is evaluated using a range of different signal-to-noise ratio values using a variety of noise types obtained from the NOISEX-92 dataset. 2017-01-01T00:00:00Z Text Conference contribution 2134/27016 https://figshare.com/articles/conference_contribution/A_comparison_of_model_validation_techniques_for_audio-visual_speech_recognition/9552131 CC BY-NC-ND 4.0 |
| spellingShingle | Mechanical engineering not elsewhere classified Audio-visual speech recognition Hidden markov models HTK toolkit Holdout validation Leave-one-out cross validation Bootstrap validation Mechanical Engineering not elsewhere classified Thum W. Seong M.Z. Ibrahim Nurul W. Arshad David Mulvaney A comparison of model validation techniques for audio-visual speech recognition |
| title | A comparison of model validation techniques for audio-visual speech recognition |
| title_full | A comparison of model validation techniques for audio-visual speech recognition |
| title_fullStr | A comparison of model validation techniques for audio-visual speech recognition |
| title_full_unstemmed | A comparison of model validation techniques for audio-visual speech recognition |
| title_short | A comparison of model validation techniques for audio-visual speech recognition |
| title_sort | comparison of model validation techniques for audio-visual speech recognition |
| topic | Mechanical engineering not elsewhere classified Audio-visual speech recognition Hidden markov models HTK toolkit Holdout validation Leave-one-out cross validation Bootstrap validation Mechanical Engineering not elsewhere classified |
| url | https://hdl.handle.net/2134/27016 |