Production-grade voice AI requires more than speech collection

Kyutai was pushing the boundaries of open voice and multimodal AI, including real-time conversational models and advanced speech generation.As the lab moved toward emotionally expressive text-to-speech, available datasets presented a limitation. Basic speech databases could support intelligibility, but they often lacked the emotional nuance, conversational variation, and multilingual consistency needed to produce natural-sounding voices.

Building the datasets introduced significant operational complexity. Speakers needed to maintain consistent identities and recording quality across sessions while delivering varied emotions and speaking styles. Transcripts, emotion labels, non-speech events, pronunciation, and metadata also had to remain accurate across languages and recording batches. To scale, Kyutai needed a governed production process capable of creating expressive multilingual data consistently.

为多语言语音数据创建统一的生产流程

Kyutai 与优步 AI 解决方案合作，将语音数据的创建流程整合为一个协同系统，而不是分别独立管理录音、转写、标注和质检。

UAIS 协调了涵盖法语、巴西葡萄牙语、欧洲和拉丁美洲西班牙语以及德语的多语言录音项目。配音演员获得了系统化的指导，旨在拓展情感和语调的表现力，包括富有同理心、讽刺、愤怒、悲伤、哭泣等高强度的说话风格。

该流程将 AI 辅助转录与人工校对相结合，用于生成逐字稿、情感标签、非语音事件及其他元数据。在所有说话人和录音批次中均采用统一的质量标准，使质检成为生产过程中的持续环节，而非最终检查点。

这种运营模式帮助 Kyutai 在项目扩展到更多语言、说话者以及日益复杂的语音需求时，始终保持了一致性。

为多语言表达型语音合成提供详尽数据集

此次合作为 Kyutai 的高表现力语音合成需求量身打造了一套结构化的多语言学习数据源。

除了数据集本身之外，此次协作还建立了一套可复用的流程，用于管理人才、记录质量、标注和验收标准，覆盖分布式语言项目。这样的基础在数据进入模型训练前就降低了差异性，并为 Kyutai 在需求变化时，持续高效地生成后续语音数据集提供了更一致的方法。