paper-with-me

홈 › Papers

IntrinsicVoice: Empowering LLMs with Intrinsic Real-time Voice Interaction Abilities

2024-10-09 · Xin Zhang, Xiang Lyu, Zhihao Du, Qian Chen, Dong Zhang, Hangrui Hu, Chaohong Tan, Tianyu Zhao, Yuxuan Wang, Bin Zhang, Heng Lu, Yaqian Zhou, Xipeng Qiu

Current methods of building LLMs with voice interaction capabilities rely heavily on explicit text autoregressive generation before or during speech response generation to maintain content quality, which unfortunately brings computational overhead and increases latency in multi-turn interactions. To address this, we introduce IntrinsicVoic,e an LLM designed with intrinsic real-time voice interaction capabilities. IntrinsicVoice aims to facilitate the transfer of textual capabilities of pre-trained LLMs to the speech modality by mitigating the modality gap between text and speech. Our novelty architecture, GroupFormer, can reduce speech sequences to lengths comparable to text sequences while generating high-quality audio, significantly reducing the length difference between speech and text, speeding up inference, and alleviating long-text modeling issues. Additionally, we construct a multi-turn speech-to-speech dialogue dataset named \method-500k which includes nearly 500k turns of speech-to-speech dialogues, and a cross-modality training strategy to enhance the semantic alignment between speech and text. Experimental results demonstrate that IntrinsicVoice can generate high-quality speech response with latency lower than 100ms in multi-turn dialogue scenarios. Demos are available at https://instrinsicvoice.github.io/.

📄 PDF Abstract BibTeX arXiv:2410.08035

Code (0)

등록된 구현이 없습니다.

Tasks

Response Generation

Similar Papers 제목 키워드 기반

Computer Environments Elicit General Agentic Intelligence in LLMs

2026-01-22 · Daixuan Cheng, Shaohan Huang, Yuxian Gu, Huatong Song 외 arxiv

Agentic intelligence in large language models (LLMs) requires not only model intrinsic capabilities but also interactions with external environments. Equipping LLMs with computers now represents a prevailing trend. Howev…

Long-Context UnderstandingInstruction Following

Position: Empowering Time Series Reasoning with Multimodal LLMs

2025-02-03 · Yaxuan Kong, Yiyuan Yang, Shiyu Wang, Chenghao Liu 외

Understanding time series data is crucial for multiple real-world applications. While large language models (LLMs) show promise in time series tasks, current approaches often rely on numerical data alone, overlooking the…

Decision MakingMultimodal ReasoningPositionTime Series+1

Empowering GraphRAG with Knowledge Filtering and Integration

2025-03-18 · Kai Guo, Harry Shomer, Shenglai Zeng, Haoyu Han 외

In recent years, large language models (LLMs) have revolutionized the field of natural language processing. However, they often suffer from knowledge gaps and hallucinations. Graph retrieval-augmented generation (GraphRA…

Retrieval-augmented Generation

GraphScout: Empowering Large Language Models with Intrinsic Exploration Ability for Agentic Graph Reasoning

2026-03-02 · Yuchen Ying, Weiqi Jiang, Tongya Zheng, Yu Wang 외 arxiv

Knowledge graphs provide structured and reliable information for many real-world applications, motivating increasing interest in combining large language models (LLMs) with graph-based retrieval to improve factual ground…

Knowledge Graphs

The Almost Intelligent Revolution: Options for Scaling Up Deliberation and Empowering People with AI

2026-06-18 · Serge Sharoff arxiv

The increasing prominence of Large Language Models (LLMs) in public discourse presents both opportunities and challenges for democratic deliberation. While red teaming strategies help mitigate specific risks, broader con…

Red Teaming