D
Inworld TTS-2
Inworld TTS-2, a text-to-speech model from DeepInfra.
speechModelOverride: { model: "inworld-tts-2" }
Inworld·served via
DeepInfra·Released 2026Pricing
- List price
- $35 / 1M characters
- Per 1M characters
- $35.00
- Per 1M characters
- $0
Specifications
- Context window
- 2,000 tokens
- Released
- 2026
- Model string
- inworld-ai/realtime-tts-2
- Capabilities
- MultilingualLow latency
Parameters
Model options, passed in
speechModelOverride.config.| Parameter | Type | Default | Range / options |
|---|---|---|---|
voicePreset voice; every voice speaks every language. Steer the delivery in the text itself, e.g. "[be happy and upbeat] Hello!". | Select | Ashley | AbbyAlainAlexAminaAnjaliArjun+121 |
temperatureHow much the delivery varies from one reading to the next. Higher is livelier, lower steadier. | Number | — | 0–2, step 0.05 |
response_format | Select | mp3 | mp3wavflacopus |
Voices · 127
Previews aren’t available for this model’s voices.
Use this model
await mindstudio.textToSpeech({ text: "...", speechModelOverride: { model: "inworld-tts-2", config: { voice: "Abby", response_format: "mp3", }, }, });