Doubao Speech
ByteDance · China · Speech & audio
China Closed source
Showing version: 豆包语音 2.0（Seed-TTS 2.0 / Seed-ICL 2.0，含 Seed-ASR 2.0） · 2025-10-16 · Version lineage ↓
01
Overview
ByteDance's Doubao speech model family covering text-to-speech, voice cloning, streaming and file transcription, and real-time voice, offered via Volcano Engine APIs; the 2.0 generation (Seed-TTS 2.0 / Seed-ASR 2.0) emphasizes LLM-based semantic understanding and emotional delivery.
Release date2025-10-16
02 / Provenance
Field checks and source notes
When sources disagree on a field, every reading is listed here with its own citation. We do not take a midpoint, invent a missing figure, or decide which side is right — follow the citation and judge for your case.
These field-level notes are recorded verbatim in Chinese in the source data. Open the Chinese version
03 / Plain layer
Plain-language benchmarks
Naturalness Not disclosed
Accuracy (WER) Not disclosed
Speed / latency Not disclosed
Languages supported Not disclosed
Price / free Not disclosed
04
Practical facts
| Context length | Not disclosed |
|---|---|
| Max output | Not disclosed |
| Parameters | Not disclosed |
| Input | Not disclosed |
| Output price | Not disclosed |
| CNY | Volcengine's Doubao page lists speech synthesis at CNY 5 per 10k characters, voice cloning at CNY 8 per 10k characters, streaming speech recognition at CNY 4.5 per hour, and file transcription 2.0 at CNY 0.8 per hour (CNY list prices, per the vendor's official pricing page). |
| Free tier | Not disclosed |
| Output speed | Not disclosed |
| Release date | |
| License | Not disclosed |
Show technical data
Technical benchmark table
No verified benchmark data (not tested / not disclosed / not applicable).
| Version | Release date | Status | Openness | Parameters | What changed | Detail | |
|---|---|---|---|---|---|---|---|
| 豆包语音 2.0（Seed-TTS 2.0 / Seed-ICL 2.0，含 Seed-ASR 2.0）Showing version | Active | Closed source | Not disclosed | TTS and voice-clone 2.0 launched 2025-10-16 with stronger emotional delivery, instruction following and formula reading; Seed-ASR 2.0 followed on 2025-12-05. | Detail | ||
| 豆包语音 1.0（语音合成 / 语音识别） | Superseded | Closed source | Not disclosed | The Doubao speech family debuted in May 2024 with TTS and ASR; official docs mark streaming ASR 1.0 as a legacy version that remains usable. | Detail |
Last updated: · Sources on detail page
06 / Compare