W
Seed Speech TTS 2.0
Seed Speech TTS 2.0, a text-to-speech model from WaveSpeed.
speechModelOverride: { model: "seed-speech-tts-2.0" }
BByteDance·served viaWaveSpeed·Released 2026
Pricing
- List price
- $60 / 1M characters
- Per generation
- $0.00010
Specifications
- Context window
- 10,000 tokens
- Released
- 2026
- Model string
- bytedance/seed-speech-tts-2.0
- Capabilities
- Multilingual
Parameters
Model options, passed in
speechModelOverride.config.| Parameter | Type | Default | Range / options |
|---|---|---|---|
voicePreset voice, named with the languages it speaks. | Select | stokie_en | vivi_mixed_en_zh_ja_es_idmindy_en_es_id_pt_zhstokie_endacey_entim_enkian_en_zh+35 |
voice_instructionOptional direction for tone, emotion, pace or volume (e.g. "warm and unhurried"). It is never spoken. | Prompt | — | — |
languageUnset, the language is detected from the text. | Select | — | zhenjaes-mxidpt-br+4 |
speedSpeaking speed. 1.0 is normal. | Number | 1 | 0.5–2, step 0.05 |
pitchPitch shift in semitones. 0 is normal. | Number | 0 | -12–12, step 1 |
volumeLoudness. 1.0 is normal. | Number | 1 | 0.5–2, step 0.05 |
output_format | Select | mp3 | mp3opus |
sample_rate | Select | 24000 | 16000240004410048000 |
Voices · 41
Previews aren’t available for this model’s voices.
Use this model
await mindstudio.textToSpeech({ text: "...", speechModelOverride: { model: "seed-speech-tts-2.0", config: { voice: "vivi_mixed_en_zh_ja_es_id", language: "zh", }, }, });