Saved in:
Bibliographic Details
Main Authors: Yuen, Robin Shing-Hei, Tse, Timothy Tin-Long, Zhu, Jian
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.17353
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912101911494656
author Yuen, Robin Shing-Hei
Tse, Timothy Tin-Long
Zhu, Jian
author_facet Yuen, Robin Shing-Hei
Tse, Timothy Tin-Long
Zhu, Jian
contents Current speech-based LLMs are predominantly trained on extensive ASR and TTS datasets, excelling in tasks related to these domains. However, their ability to handle direct speech-to-speech conversations remains notably constrained. These models often rely on an ASR-to-TTS chain-of-thought pipeline, converting speech into text for processing before generating audio responses, which introduces latency and loses audio features. We propose a method that implicitly internalizes ASR chain of thought into a speech LLM, enhancing its native speech understanding capabilities. Our approach reduces latency and improves the model's native understanding of speech, paving the way for more efficient and natural real-time audio interactions. We also release a large-scale synthetic conversational dataset to facilitate further research.
format Preprint
id arxiv_https___arxiv_org_abs_2409_17353
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Internalizing ASR with Implicit Chain of Thought for Efficient Speech-to-Speech Conversational LLM
Yuen, Robin Shing-Hei
Tse, Timothy Tin-Long
Zhu, Jian
Computation and Language
Current speech-based LLMs are predominantly trained on extensive ASR and TTS datasets, excelling in tasks related to these domains. However, their ability to handle direct speech-to-speech conversations remains notably constrained. These models often rely on an ASR-to-TTS chain-of-thought pipeline, converting speech into text for processing before generating audio responses, which introduces latency and loses audio features. We propose a method that implicitly internalizes ASR chain of thought into a speech LLM, enhancing its native speech understanding capabilities. Our approach reduces latency and improves the model's native understanding of speech, paving the way for more efficient and natural real-time audio interactions. We also release a large-scale synthetic conversational dataset to facilitate further research.
title Internalizing ASR with Implicit Chain of Thought for Efficient Speech-to-Speech Conversational LLM
topic Computation and Language
url https://arxiv.org/abs/2409.17353