Enhancing Generalization of Speech Large Language Models with Multi-Task Behavior Imitation and Speech-Text Interleaving

• URL: http://arxiv.org/abs/2505.18644v1
• Date: Sat, 24 May 2025 11:09:13 GMT
• Title: Enhancing Generalization of Speech Large Language Models with Multi-Task Behavior Imitation and Speech-Text Interleaving
• Authors: Jingran Xie, Xiang Li, Hui Wang, Yue Yu, Yang Xiang, Xixin Wu, Zhiyong Wu,
• Abstract summary: Large language models (LLMs) have shown remarkable generalization across tasks.<n>LLMs typically use supervised fine-tuning to align speech with text-based LLMs.<n>We propose a novel multi-task 'behavior imitation' method with speech-text interleaving.
• Score: 36.246791887458194
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) have shown remarkable generalization across tasks, leading to increased interest in integrating speech with LLMs. These speech LLMs (SLLMs) typically use supervised fine-tuning to align speech with text-based LLMs. However, the lack of annotated speech data across a wide range of tasks hinders alignment efficiency, resulting in poor generalization. To address these issues, we propose a novel multi-task 'behavior imitation' method with speech-text interleaving, called MTBI, which relies solely on paired speech and transcripts. By ensuring the LLM decoder generates equivalent responses to paired speech and text, we achieve a more generalized SLLM. Interleaving is used to further enhance alignment efficiency. We introduce a simple benchmark to evaluate prompt and task generalization across different models. Experimental results demonstrate that our MTBI outperforms SOTA SLLMs on both prompt and task generalization, while requiring less supervised speech data.

Related papers

• A Unified Speech LLM for Diarization and Speech Recognition in Multilingual Conversations [25.58593495281218]
  We propose a unified speech LLM that jointly performs diarization and ASR in an end-to-end manner.<n>By reformulating the training data format and modifying the inference procedure, our model addresses the ambiguity inherent in pre-segmented audio.
  arXiv  Detail & Related papers  (2025-06-26T01:54:02Z)
• What Makes a Good Speech Tokenizer for LLM-Centric Speech Generation? A Systematic Study [58.55905182336196]
  Speech-language models (SLMs) offer a promising path toward unifying speech and text understanding and generation.<n>We investigate the role of speech tokenizer designs in LLM-centric SLMs, augmented by speech heads and speaker modeling.<n>We introduce multi-token prediction (MTP) into SLMs, enabling each hidden state to decode multiple speech tokens.
  arXiv  Detail & Related papers  (2025-06-14T15:26:31Z)
• Get Large Language Models Ready to Speak: A Late-fusion Approach for Speech Generation [14.746190461312036]
  Large language models (LLMs) have revolutionized natural language processing (NLP) We introduce a text-to-speech (TTS) system powered by a fine-tuned Llama model, named TTS-Llama, that achieves state-of-the-art speech synthesis performance. We further propose MoLE-Llama, a text-and-speech multimodal LLM developed through purely late-fusion parameter-efficient fine-tuning (PEFT) and a mixture-of-expert architecture.
  arXiv  Detail & Related papers  (2024-10-27T04:28:57Z)
• VoiceTextBlender: Augmenting Large Language Models with Speech Capabilities via Single-Stage Joint Speech-Text Supervised Fine-Tuning [64.56272011710735]
  We propose a novel single-stage joint speech-text SFT approach on the low-rank adaptation (LoRA) of the large language models (LLMs) backbone.<n>Compared to previous SpeechLMs with 7B or 13B parameters, our 3B model demonstrates superior performance across various speech benchmarks.
  arXiv  Detail & Related papers  (2024-10-23T00:36:06Z)
• Self-Powered LLM Modality Expansion for Large Speech-Text Models [62.27700381806554]
  Large language models (LLMs) exhibit remarkable performance across diverse tasks. This study aims to refine the use of speech datasets for LSM training by addressing the limitations of vanilla instruction tuning. We introduce a self-powered LSM that leverages augmented automatic speech recognition data generated by the model itself for more effective instruction tuning.
  arXiv  Detail & Related papers  (2024-10-04T04:34:24Z)
• DeSTA2: Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning Data [84.01401439030265]
  Recent end-to-end speech language models (SLMs) have expanded upon the capabilities of large language models (LLMs)<n>We present a simple yet effective automatic process for creating speech-text pair data.<n>Our model demonstrates general capabilities for speech-related tasks without the need for speech instruction-tuning data.
  arXiv  Detail & Related papers  (2024-09-30T07:01:21Z)
• Large Language Model Can Transcribe Speech in Multi-Talker Scenarios with Versatile Instructions [68.98811048970963]
  We present a pioneering effort to investigate the capability of large language models (LLMs) in transcribing speech in multi-talker environments.<n>We use WavLM and Whisper encoder to extract multi-faceted speech representations that are sensitive to speaker characteristics and semantic context.<n>Experiments reveal the promising performance of our proposed system, MT-LLM, in cocktail party scenarios.
  arXiv  Detail & Related papers  (2024-09-13T07:28:28Z)
• SpeechGen: Unlocking the Generative Power of Speech Language Models with Prompts [108.04306136086807]
  We present research that explores the application of prompt tuning to stimulate speech LMs for various generation tasks, within a unified framework called SpeechGen. The proposed unified framework holds great promise for efficiency and effectiveness, particularly with the imminent arrival of advanced speech LMs.
  arXiv  Detail & Related papers  (2023-06-03T22:35:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.