End-to-end audiovisual speech recognition
Several end-to-end deep learning approaches have been recently presented which extract either audio or visual features from the input images or audio signals and perform speech recognition. However, research on end-to-end audiovisual models is very limited. In this work, we present an end-to-end aud...
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2018
|
| Online Access: | https://eprints.nottingham.ac.uk/51132/ |
| _version_ | 1848798424187535360 |
|---|---|
| author | Petridis, Stavros Stafylakis, Themos Ma, Pingchuan Cai, Feipeng Tzimiropoulos, Georgios Pantic, Maja |
| author_facet | Petridis, Stavros Stafylakis, Themos Ma, Pingchuan Cai, Feipeng Tzimiropoulos, Georgios Pantic, Maja |
| author_sort | Petridis, Stavros |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Several end-to-end deep learning approaches have been recently presented which extract either audio or visual features from the input images or audio signals and perform speech recognition. However, research on end-to-end audiovisual models is very limited. In this work, we present an end-to-end audiovisual model based on residual networks and Bidirectional Gated Recurrent Units (BGRUs). To the best of our knowledge, this is the first audiovisual fusion model which simultaneously learns to extract features directly from the image pixels and audio waveforms and performs within-context word recognition on a large publicly available dataset (LRW). The model consists of two streams, one for each modality, which extract features directly from mouth regions and raw waveforms. The temporal dynamics in each stream/modality are modeled by a 2-layer BGRU and the fusion of multiple streams/modalities takes place via another 2-layer BGRU. A slight improvement in the classification rate over an end-to-end audio-only and MFCC-based model is reported in clean audio conditions and low levels of noise. In presence of high levels of noise, the end-to-end audiovisual model significantly outperforms both audio-only models. |
| first_indexed | 2025-11-14T20:19:33Z |
| format | Conference or Workshop Item |
| id | nottingham-51132 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:19:33Z |
| publishDate | 2018 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-511322018-04-15T04:35:06Z https://eprints.nottingham.ac.uk/51132/ End-to-end audiovisual speech recognition Petridis, Stavros Stafylakis, Themos Ma, Pingchuan Cai, Feipeng Tzimiropoulos, Georgios Pantic, Maja Several end-to-end deep learning approaches have been recently presented which extract either audio or visual features from the input images or audio signals and perform speech recognition. However, research on end-to-end audiovisual models is very limited. In this work, we present an end-to-end audiovisual model based on residual networks and Bidirectional Gated Recurrent Units (BGRUs). To the best of our knowledge, this is the first audiovisual fusion model which simultaneously learns to extract features directly from the image pixels and audio waveforms and performs within-context word recognition on a large publicly available dataset (LRW). The model consists of two streams, one for each modality, which extract features directly from mouth regions and raw waveforms. The temporal dynamics in each stream/modality are modeled by a 2-layer BGRU and the fusion of multiple streams/modalities takes place via another 2-layer BGRU. A slight improvement in the classification rate over an end-to-end audio-only and MFCC-based model is reported in clean audio conditions and low levels of noise. In presence of high levels of noise, the end-to-end audiovisual model significantly outperforms both audio-only models. 2018-04-15 Conference or Workshop Item PeerReviewed application/pdf en https://eprints.nottingham.ac.uk/51132/1/av_speech1.pdf Petridis, Stavros, Stafylakis, Themos, Ma, Pingchuan, Cai, Feipeng, Tzimiropoulos, Georgios and Pantic, Maja (2018) End-to-end audiovisual speech recognition. In: IEEE International Conference on Acoustics, Speech, and Signal Processing, 15-20 April 2018, Calgary, Alberta, Canada. |
| spellingShingle | Petridis, Stavros Stafylakis, Themos Ma, Pingchuan Cai, Feipeng Tzimiropoulos, Georgios Pantic, Maja End-to-end audiovisual speech recognition |
| title | End-to-end audiovisual speech recognition |
| title_full | End-to-end audiovisual speech recognition |
| title_fullStr | End-to-end audiovisual speech recognition |
| title_full_unstemmed | End-to-end audiovisual speech recognition |
| title_short | End-to-end audiovisual speech recognition |
| title_sort | end-to-end audiovisual speech recognition |
| url | https://eprints.nottingham.ac.uk/51132/ |