Chunked Attention-based Encoder-Decoder Model for Streaming Speech Recognition
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929211846950912 |
|---|---|
| author | Zeineldeen, Mohammad Zeyer, Albert Schlüter, Ralf Ney, Hermann |
| author_facet | Zeineldeen, Mohammad Zeyer, Albert Schlüter, Ralf Ney, Hermann |
| contents | We study a streamable attention-based encoder-decoder model in which either the decoder, or both the encoder and decoder, operate on pre-defined, fixed-size windows called chunks. A special end-of-chunk (EOC) symbol advances from one chunk to the next chunk, effectively replacing the conventional end-of-sequence symbol. This modification, while minor, situates our model as equivalent to a transducer model that operates on chunks instead of frames, where EOC corresponds to the blank symbol. We further explore the remaining differences between a standard transducer and our model. Additionally, we examine relevant aspects such as long-form speech generalization, beam size, and length normalization. Through experiments on Librispeech and TED-LIUM-v2, and by concatenating consecutive sequences for long-form trials, we find that our streamable model maintains competitive performance compared to the non-streamable variant and generalizes very well to long-form speech. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2309_08436 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Chunked Attention-based Encoder-Decoder Model for Streaming Speech Recognition Zeineldeen, Mohammad Zeyer, Albert Schlüter, Ralf Ney, Hermann Audio and Speech Processing Sound Machine Learning We study a streamable attention-based encoder-decoder model in which either the decoder, or both the encoder and decoder, operate on pre-defined, fixed-size windows called chunks. A special end-of-chunk (EOC) symbol advances from one chunk to the next chunk, effectively replacing the conventional end-of-sequence symbol. This modification, while minor, situates our model as equivalent to a transducer model that operates on chunks instead of frames, where EOC corresponds to the blank symbol. We further explore the remaining differences between a standard transducer and our model. Additionally, we examine relevant aspects such as long-form speech generalization, beam size, and length normalization. Through experiments on Librispeech and TED-LIUM-v2, and by concatenating consecutive sequences for long-form trials, we find that our streamable model maintains competitive performance compared to the non-streamable variant and generalizes very well to long-form speech. |
| title | Chunked Attention-based Encoder-Decoder Model for Streaming Speech Recognition |
| topic | Audio and Speech Processing Sound Machine Learning |
| url | https://arxiv.org/abs/2309.08436 |