Transducer-Llama: Integrating LLMs into Streamable Transducer-based Speech Recognition
This work addresses the problem of making LLMs streamable for speech recognition, which is incremental as it builds on existing transducer and LLM methods.
The paper tackled the challenge of integrating large language models (LLMs) into streamable speech recognition by proposing Transducer-Llama, which combines LLMs with a Factorized Transducer model and uses a weak-to-strong LM swap strategy, achieving a 17% relative WER reduction over a strong baseline and 32% over an RNN-T baseline.
While large language models (LLMs) have been applied to automatic speech recognition (ASR), the task of making the model streamable remains a challenge. This paper proposes a novel model architecture, Transducer-Llama, that integrates LLMs into a Factorized Transducer (FT) model, naturally enabling streaming capabilities. Furthermore, given that the large vocabulary of LLMs can cause data sparsity issue and increased training costs for spoken language systems, this paper introduces an efficient vocabulary adaptation technique to align LLMs with speech system vocabularies. The results show that directly optimizing the FT model with a strong pre-trained LLM-based predictor using the RNN-T loss yields some but limited improvements over a smaller pre-trained LM predictor. Therefore, this paper proposes a weak-to-strong LM swap strategy, using a weak LM predictor during RNN-T loss training and then replacing it with a strong LLM. After LM replacement, the minimum word error rate (MWER) loss is employed to finetune the integration of the LLM predictor with the Transducer-Llama model. Experiments on the LibriSpeech and large-scale multi-lingual LibriSpeech corpora show that the proposed streaming Transducer-Llama approach gave a 17% relative WER reduction (WERR) over a strong FT baseline and a 32% WERR over an RNN-T baseline.