All Models
Seed Speech TTS 2.0
Seed Speech TTS 2.0 converts text into natural speech with multilingual voices and MP3 or Opus output.
Call the API endpoint to get the latest, detailed pricing and pricing estimates.
Call the API endpoint to get the latest, detailed pricing and pricing estimates.
Call the API endpoint to get the latest, detailed pricing and pricing estimates.
Seed Speech TTS 2.0
Seed Speech TTS 2.0 is ByteDance Seed’s text-to-speech model. You pick a named catalog voice and a language, then pass a script. Default voice is Stokie in English. Other named voices include Dacey and mixed multilingual casts.
Languages are Chinese, English, Japanese, Mexican Spanish, Indonesian, Brazilian Portuguese, Korean, Italian, German, and French. Pitch runs from −12 to 12. Speed and volume run from 0.5 to 2. Sample rates go from 8 kHz to 48 kHz, default 24 kHz. Output is MP3 or Opus. An optional voice instruction field is there for delivery notes.
It is speech, not music, and not lip-sync. There is no clone-from-a-recording path; you choose from the roster.
What it's good for
Product narration, VO beds, and any library of clips that has to stay on the same named voice take after take. Match the language setting to the script. Mixed voices exist for bilingual lines.
Model ID
bytedance/seed-speech-tts-2.0
Output Type
Audio
Pricing
$0.0001
Provider
ByteDance
Hosting
Protoface
Seed Speech TTS 2.0
Seed Speech TTS 2.0 is ByteDance Seed’s text-to-speech model. You pick a named catalog voice and a language, then pass a script. Default voice is Stokie in English. Other named voices include Dacey and mixed multilingual casts.
Languages are Chinese, English, Japanese, Mexican Spanish, Indonesian, Brazilian Portuguese, Korean, Italian, German, and French. Pitch runs from −12 to 12. Speed and volume run from 0.5 to 2. Sample rates go from 8 kHz to 48 kHz, default 24 kHz. Output is MP3 or Opus. An optional voice instruction field is there for delivery notes.
It is speech, not music, and not lip-sync. There is no clone-from-a-recording path; you choose from the roster.
What it's good for
Product narration, VO beds, and any library of clips that has to stay on the same named voice take after take. Match the language setting to the script. Mixed voices exist for bilingual lines.