Using Large Language Model for End-to-End Chinese ASR and NER
This work addresses speech-text integration for Chinese ASR and NER, showing incremental improvements in error reduction and SOTA performance on a specific benchmark.
The paper tackled the problem of integrating speech into large language models for Chinese automatic speech recognition and name entity recognition, comparing encoder-decoder and decoder-only architectures, and achieved a state-of-the-art F1 score of 0.805 on the AISHELL-NER test set using chain-of-thought NER.
Mapping speech tokens to the same feature space as text tokens has become the paradigm for the integration of speech modality into decoder-only large language models (LLMs). An alternative approach is to use an encoder-decoder architecture that incorporates speech features through cross-attention. This approach, however, has received less attention in the literature. In this work, we connect the Whisper encoder with ChatGLM3 and provide in-depth comparisons of these two approaches using Chinese automatic speech recognition (ASR) and name entity recognition (NER) tasks. We evaluate them not only by conventional metrics like the F1 score but also by a novel fine-grained taxonomy of ASR-NER errors. Our experiments reveal that encoder-decoder architecture outperforms decoder-only architecture with a short context, while decoder-only architecture benefits from a long context as it fully exploits all layers of the LLM. By using LLM, we significantly reduced the entity omission errors and improved the entity ASR accuracy compared to the Conformer baseline. Additionally, we obtained a state-of-the-art (SOTA) F1 score of 0.805 on the AISHELL-NER test set by using chain-of-thought (CoT) NER which first infers long-form ASR transcriptions and then predicts NER labels.