IndexTeam

Open-source speech models for controllable zero-shot TTS

IndexTeam maintains the public IndexTTS repositories and model releases for zero-shot neural text-to-speech. The project pages identify IndexTeam and Bilibili as developers of IndexTTS-2.5, with model weights, demos, technical reports, and installation instructions published for builders. The group focuses on speech synthesis systems that can clone a voice from a short reference clip, transfer voices across languages, and control emotion, speaking speed, and pronunciation. Its work is relevant to developers building dubbing, localization, accessibility, and voice-agent workflows. OpenTools tracks IndexTeam as the organization dependency for IndexTTS so model pages can link back to the maintaining team and avoid orphan LLM/model records.