paper-with-me

홈 › Papers

Globally Normalising the Transducer for Streaming Speech Recognition

2023-07-20 · Rogier Van Dalen

The Transducer (e.g. RNN-Transducer or Conformer-Transducer) generates an output label sequence as it traverses the input sequence. It is straightforward to use in streaming mode, where it generates partial hypotheses before the complete input has been seen. This makes it popular in speech recognition. However, in streaming mode the Transducer has a mathematical flaw which, simply put, restricts the model's ability to change its mind. The fix is to replace local normalisation (e.g. a softmax) with global normalisation, but then the loss function becomes impossible to evaluate exactly. A recent paper proposes to solve this by approximating the model, severely degrading performance. Instead, this paper proposes to approximate the loss function, allowing global normalisation to apply to a state-of-the-art streaming model. Global normalisation reduces its word error rate by 9-11% relative, closing almost half the gap between streaming and lookahead mode.

📄 PDF Abstract BibTeX arXiv:2307.10975

Code (0)

등록된 구현이 없습니다.

Tasks

speech-recognitionSpeech Recognition

Similar Papers 제목 키워드 기반

Global Normalization for Streaming Speech Recognition in a Modular Framework

2022-05-26 · Ehsan Variani, Ke wu, Michael Riley, David Rybach 외

We introduce the Globally Normalized Autoregressive Transducer (GNAT) for addressing the label bias problem in streaming speech recognition. Our solution admits a tractable exact computation of the denominator for the se…

speech-recognitionSpeech Recognition

LAMASSU: Streaming Language-Agnostic Multilingual Speech Recognition and Translation Using Neural Transducers

2022-11-05 · Peidong Wang, Eric Sun, Jian Xue, Yu Wu 외

Automatic speech recognition (ASR) and speech translation (ST) can both use neural transducers as the model structure. It is thus possible to use a single transducer model to perform both tasks. In real-world application…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Language Identificationspeech-recognition+3

Towards Maximum Likelihood Training for Transducer-based Streaming Speech Recognition

2024-11-26 · Hyeonseung Lee, Ji Won Yoon, Sungsoo Kim, Nam Soo Kim

Transducer neural networks have emerged as the mainstream approach for streaming automatic speech recognition (ASR), offering state-of-the-art performance in balancing accuracy and latency. In the conventional framework,…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

Improving Fast-slow Encoder based Transducer with Streaming Deliberation

2022-12-15 · Ke Li, Jay Mahadeokar, Jinxi Guo, Yangyang Shi 외

This paper introduces a fast-slow encoder based transducer with streaming deliberation for end-to-end automatic speech recognition. We aim to improve the recognition accuracy of the fast-slow encoder based transducer whi…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition+1

Speech-T: Transducer for Text to Speech and Beyond

2021-12-01 · NeurIPS 2021 12 · Jiawei Chen, Xu Tan, Yichong Leng, Jin Xu 외

Neural Transducer (e.g., RNN-T) has been widely used in automatic speech recognition (ASR) due to its capabilities of efficiently modeling monotonic alignments between input and output sequences and naturally supporting …

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Sentencespeech-recognition+3