Seed-TTS 2.0
ByteDance's streaming TTS model with bidirectional realtime synthesis on Volcengine.
About
Seed-TTS 2.0 is ByteDance's production text-to-speech model served through the Volcengine speech platform and the Doubao API. It offers bidirectional and unidirectional streaming endpoints that convert LLM token streams into audio in realtime, plus HTTP chunked and SSE variants for flexible integration.
The model family targets conversational agents and realtime applications where low-latency, high-quality synthesis matters, with zero-shot voice capabilities and emotion control highlighted in community integrations. Seed-TTS 2.0 is a closed API model; it is not released as open weights, partly due to AI safety concerns around voice cloning.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 106; Trending score 78/100 from researcher assessment.