CUSIDE-T: Chunking, Simulating Future and Decoding for Transducer based Streaming ASR
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914949800919040 |
|---|---|
| author | Zhao, Wenbo Li, Ziwei Yu, Chuan Ou, Zhijian |
| author_facet | Zhao, Wenbo Li, Ziwei Yu, Chuan Ou, Zhijian |
| contents | Streaming automatic speech recognition (ASR) is very important for many real-world ASR applications. However, a notable challenge for streaming ASR systems lies in balancing operational performance against latency constraint. Recently, a method of chunking, simulating future context and decoding, called CUSIDE, has been proposed for connectionist temporal classification (CTC) based streaming ASR, which obtains a good balance between reduced latency and high recognition accuracy. In this paper, we present CUSIDE-T, which successfully adapts the CUSIDE method over the recurrent neural network transducer (RNN-T) ASR architecture, instead of being based on the CTC architecture. We also incorporate language model rescoring in CUSIDE-T to further enhance accuracy, while only bringing a small additional latency. Extensive experiments are conducted over the AISHELL-1, WenetSpeech and SpeechIO datasets, comparing CUSIDE-T and U2++ (both based on RNN-T). U2++ is an existing counterpart of chunk based streaming ASR method. It is shown that CUSIDE-T achieves superior accuracy performance for streaming ASR, with equal settings of latency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_10255 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CUSIDE-T: Chunking, Simulating Future and Decoding for Transducer based Streaming ASR Zhao, Wenbo Li, Ziwei Yu, Chuan Ou, Zhijian Sound Audio and Speech Processing Streaming automatic speech recognition (ASR) is very important for many real-world ASR applications. However, a notable challenge for streaming ASR systems lies in balancing operational performance against latency constraint. Recently, a method of chunking, simulating future context and decoding, called CUSIDE, has been proposed for connectionist temporal classification (CTC) based streaming ASR, which obtains a good balance between reduced latency and high recognition accuracy. In this paper, we present CUSIDE-T, which successfully adapts the CUSIDE method over the recurrent neural network transducer (RNN-T) ASR architecture, instead of being based on the CTC architecture. We also incorporate language model rescoring in CUSIDE-T to further enhance accuracy, while only bringing a small additional latency. Extensive experiments are conducted over the AISHELL-1, WenetSpeech and SpeechIO datasets, comparing CUSIDE-T and U2++ (both based on RNN-T). U2++ is an existing counterpart of chunk based streaming ASR method. It is shown that CUSIDE-T achieves superior accuracy performance for streaming ASR, with equal settings of latency. |
| title | CUSIDE-T: Chunking, Simulating Future and Decoding for Transducer based Streaming ASR |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2407.10255 |