Advanced Long-Content Speech Recognition With Factorized Neural Transducer

• Published in: IEEE/ACM transactions on audio, speech, and language processing speech, and language processing, 2024, Vol.32, p.1803-1815
• Main Authors: Gong, Xun, Wu, Yu, Li, Jinyu, Liu, Shujie, Zhao, Rui, Chen, Xie, Qian, Yanmin
• Format: Article
• Language: English
• Subjects: Acoustics; Automatic speech recognition; Computer architecture; Context modeling; Error analysis; Error reduction; Factorized neural transducer; Language modeling; long-content speech recognition; Network latency; Predictive models; RNN-T; Speech; Speech recognition; streaming and non-streaming; Transcription; Transducers; Transformers; Vocabulary; Voice recognition
• ISSN: 2329-9290; 2329-9304
• DOI: 10.1109/TASLP.2024.3350893

Long-content automatic speech recognition (ASR) has obtained increasing interest in recent years, as it captures the relationship among consecutive historical utterances while decoding the current utterance. In this paper, we propose two novel approaches, which integrate long-content information into the factorized neural transducer (FNT) based architecture in both non-streaming (referred to as LongFNT ) and streaming (referred to as SLongFNT ) scenarios. We first investigate whether long-content transcriptions can improve the vanilla conformer transducer (C-T) models. Our experiments indicate that the vanilla C-T models do not exhibit improved performance when utilizing long-content transcriptions, possibly due to the predictor network of C-T models not functioning as a pure language model. Instead, FNT shows its potential in utilizing long-content information, where we propose the LongFNT model and explore the impact of long-content information in both text (LongFNT-Text) and speech (LongFNT-Speech). The proposed LongFNT-Text and LongFNT-Speech models further complement each other to achieve better performance, with transcription history proving more valuable to the model. The effectiveness of our LongFNT approach is evaluated on LibriSpeech and GigaSpeech corpora, and obtains relative 19% and 12% word error rate reduction, respectively. Furthermore, we extend the LongFNT model to the streaming scenario, which is named SLongFNT , consisting of SLongFNT-Text and SLongFNT-Speech approaches to utilize long-content text and speech information. Experiments show that the proposed SLongFNT model achieves relative 26% and 17% WER reduction on LibriSpeech and GigaSpeech respectively while keeping a good latency, compared to the FNT baseline. Overall, our proposed LongFNT and SLongFNT highlight the significance of considering long-content speech and transcription knowledge for improving both non-streaming and streaming speech recognition systems.