用语音标记语言提升法语合成语音的自然度。
Improving French Synthetic Speech Quality via SSML Prosody Control
- 通过两个微调模型自动插入SSML标签控制语调、语速等参数。
- 断句准确率达99.2%，语调等误差降低25%-40%。
- 听觉评测显示自然度显著提升，15人中14人更偏好该方法。
尽管近期取得进展，商业文本转语音系统仍因语调控制有限而缺乏表现力。本文提出首个端到端流程，将语音合成标记语言（SSML）标签插入法语文本，以控制音高、语速、音量和停顿时长。采用两级级联架构，使用两个微调的Qwen 2.5-7B模型：第一个预测短语断点位置，第二个对语调目标进行回归，生成兼容商业TTS系统的SSML标记。在14小时法语播客语料库上评估，断点识别F1达99.2%，音高、语速和音量的平均绝对误差比仅用提示的大型语言模型（LLMs）和BiLSTM基线降低25%-40%。在涉及18名参与者、超过9小时合成音频的感知评测中，使用SSML增强的语音显著提升自然度，平均意见得分从3.20升至3.87（p < 0.005）。此外，18人中有15人更偏好该合成结果。这些结果表明，该方法在缩小合成语音与自然语音之间的表现力差距方面取得显著进展。代码已公开于https://github.com/hi-paris/Prosody-Control-French-TTS。
原文摘要 · Abstract (English)
Despite recent advances, synthetic voices often lack expressiveness due to limited prosody control in commercial text-to-speech (TTS) systems. We introduce the first end-to-end pipeline that inserts Speech Synthesis Markup Language (SSML) tags into French text to control pitch, speaking rate, volume, and pause duration. We employ a cascaded architecture with two QLoRA-fine-tuned Qwen 2.5-7B models: one predicts phrase-break positions and the other performs regression on prosodic targets, generating commercial TTS-compatible SSML markup. Evaluated on a 14-hour French podcast corpus, our method achieves 99.2% F1 for break placement and reduces mean absolute error on pitch, rate, and volume by 25-40% compared with prompting-only large language models (LLMs) and a BiLSTM baseline. In perceptual evaluation involving 18 participants across over 9 hours of synthesized audio, SSML-enhanced speech generated by our pipeline significantly improves naturalness, with the mean opinion score increasing from 3.20 to 3.87 (p < 0.005). Additionally, 15 of 18 listeners preferred our enhanced synthesis. These results demonstrate substantial progress in bridging the expressiveness gap between synthetic and natural French speech. Our code is publicly available at https://github.com/hi-paris/Prosody-Control-French-TTS.
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。