Fastemit: Low-latency Streaming ASR With Sequence-level Emission Regularization

Abstract

Streaming automatic speech recognition (ASR) aims to emit each hypothesized word as quickly and accurately as possible. However, emitting fast without degrading quality, as measured by word error rate (WER), is highly challenging. Existing approaches including Early and Late Penalties and Constrained Alignments penalize emission delay by manipulating per-token or per-frame probability prediction in sequence transducer models. While being successful in reducing delay, these approaches suffer from significant accuracy regression and also require additional word alignment information from an existing model. In this work, we propose a sequence-level emission regularization method, named FastEmit, that applies latency regularization directly on per-sequence probability in training transducer models, and does not require any alignment. We demonstrate that FastEmit is more suitable to the sequence-level optimization of transducer models for streaming ASR by applying it on various end-to-end s

Stats

• citations47
• S2 citations—
• github stars0
• HF likes0
• heat score12.61
• arxiv keyyu2020fastemit

Related papers

• Stableemit: Selection Probability Discount For Reducing Emission Latency Of Streaming Monotonic Attention ASR (2021)3.58
• Reducing Streaming ASR Model Delay With Self Alignment (2021)6.77
• Streaming Parallel Transducer Beam Search With Fast-slow Cascaded Encoders (2022)0.00
• Minimum Latency Training Of Sequence Transducers For Streaming End-to-end Speech Recognition (2022)0.00
• High Performance Sequence-to-sequence Model For Streaming Speech Recognition (2020)3.58
• Extreme Encoder Output Frame Rate Reduction: Improving Computational Latencies Of Large End-to-end Models (2024)5.84
• Minimum Latency Training Strategies For Streaming Sequence-to-sequence ASR (2020)10.07
• Dynamic Latency For Ctc-based Streaming Automatic Speech Recognition With Emformer (2022)0.00