Providers
Inworld
Inworld text-to-speech through the Speechbase gateway, with streaming, multilingual voices, and native timestamps.
| Prefix | inworld |
| Default model | inworld-tts-1.5-max |
| Provider key | Connect under Provider Keys |
Route to Inworld by prefixing the model with inworld/. All models stream and
return native word-level timestamps.
Models
| Model | Streaming | Timestamps | Max input | Languages |
|---|---|---|---|---|
inworld-tts-2 | Yes | Native | 2000 | 14 |
inworld-tts-1.5-max | Yes | Native | 2000 | 14 |
inworld-tts-1.5-mini | Yes | Native | 2000 | 14 |
inworld-tts-2 is the newest; the 1.5-mini model is the lowest-latency option.
Usage
curl -X POST https://api.speechbase.ai/v1/audio/speech \
-H "Authorization: Bearer $SPEECHBASE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"mode": "inline",
"model": "inworld/inworld-tts-1.5-max",
"voice": "Ashley",
"text": "Hello from Speechbase!",
"output": "mp3"
}' --output hello.mp3voice is an Inworld voice name such as Ashley, Dominus, Edward, Hades,
or Priya.
Provider options
Anything in providerOptions is forwarded to the Inworld API unchanged.