用大模型生成语音质量标签，解决标注数据少的问题。
Leveraging LLMs for Scalable Non-intrusive Speech Quality Assessment
- 用微调后的大模型作为伪评分员生成语音质量标签
- 两阶段训练使跨场景评估性能提升，最高相关系数达0.73
- 适合需要低成本扩展语音质量评估能力的研究者
非侵入式语音质量评估（SQA）系统受限于训练数据不足和人工标注成本高，难以泛化到实时会议场景。本文提出利用大语言模型（LLM）作为伪评分员生成语音质量标签，以缓解数据瓶颈。构建了包含101,129段带模拟退化的语音片段的LibriAugmented数据集，由微调后的听觉大模型Vicuna-7b-v1.5进行标注。对比了三种训练策略：仅使用人工标注、仅使用LLM标注、两阶段方法（先用LLM标签预训练，再用人工标签微调），采用DNSMOS Pro和DeePMOS作为评估指标。在多个跨语言、跨退化类型的测试集上验证。尽管仅用LLM标签训练效果参差不齐，但两阶段方法显著提升了泛化性能：在NISQA_TEST_LIVETALK上，DNSMOS Pro的PCC从0.55提升至0.63；在腾讯含混响数据集上，从0.65提升至0.73。结果证明，利用大模型作为可扩展伪评分员具有潜力，为解决数据限制问题提供了低成本方案。
原文摘要 · Abstract (English)
Non-intrusive speech quality assessment (SQA) systems suffer from limited training data and costly human annotations, hindering their generalization to real-time conferencing calls. In this work, we propose leveraging large language models (LLMs) as pseudo-raters for speech quality to address these data bottlenecks. We construct LibriAugmented, a dataset consisting of 101,129 speech clips with simulated degradations labeled by a fine-tuned auditory LLM (Vicuna-7b-v1.5). We compare three training strategies: using human-labeled data, using LLM-labeled data, and a two-stage approach (pretraining on LLM labels, then fine-tuning on human labels), using both DNSMOS Pro and DeePMOS. We test on several datasets across languages and quality degradations. While LLM-labeled training yields mixed results compared to human-labeled training, we provide empirical evidence that the two-stage approach improves the generalization performance (e.g., DNSMOS Pro achieves 0.63 vs. 0.55 PCC on NISQA_TEST_LIVETALK and 0.73 vs. 0.65 PCC on Tencent with reverb). Our findings demonstrate the potential of using LLMs as scalable pseudo-raters for speech quality assessment, offering a cost-effective solution to the data limitation problem.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。