Seed TTS 2.0
Seed TTS 2.0 uses a query-and-response design to account for conversational context when producing speech. Text instructions can control tone, emotion and pace.
Technical guide
The interface can use earlier context to shape delivery and sentence-level tags to guide emotion. The listed languages include English, Chinese, Japanese, Korean, Arabic and several European and Southeast Asian languages.Seed TTS 2.0 overview
Practical use
Practical guidance. Supply the conversation context only when it changes how a line should sound. Listen for accidental changes to names and numbers as well as the requested emotional delivery.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | seed-tts-2.0 |
| API selection | Use resource ID seed-tts-2.0.Seed speech synthesis and voice replication API |
| Delivery | HTTP streaming and WebSocket modes are documented.Seed TTS 2.0 overview |
Related models
- Seed Audio 1.0
ByteDance · Seed Audio
Speech, sound-effect and ambient-audio generator combining language-based control with acoustic diffusion.
Launch date not confirmedAPI - Seed Voice Replication 2.0
ByteDance · Seed Speech
Voice replication model that learns a reusable voice profile from a short single-speaker recording.
Launch date not confirmedAPI