curl --request POST \
--url https://api.slng.ai/v1/bridges/unmute/tts/{model_variant} \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"voice": "aura-2-thalia-en",
"text": "Hello, this is a test of text to speech synthesis."
}
'"<string>"{
"error": "Validation error",
"details": "Missing required field: text"
}{
"error": "TTS service error",
"status": 400,
"details": "Upstream provider error: 400 Bad Request - Invalid voice"
}{
"error": "<string>",
"message": "<string>",
"statusCode": 123
}{
"error": "Service error",
"details": "Internal processing error"
}{
"error": "TTS service error",
"status": 400,
"details": "Upstream provider error: 400 Bad Request - Invalid voice"
}Unified TTS
Unified TTS
Synthesize speech through the Unmute bridge over HTTP. The model_variant path parameter selects the target TTS model (e.g., deepgram/aura:2), and the gateway maps voice, text, and language to that model’s HTTP request fields. Models that only stream over SSE return 400 on this route; use the WebSocket channel instead.
POST
/
v1
/
bridges
/
unmute
/
tts
/
{model_variant}
curl --request POST \
--url https://api.slng.ai/v1/bridges/unmute/tts/{model_variant} \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"voice": "aura-2-thalia-en",
"text": "Hello, this is a test of text to speech synthesis."
}
'"<string>"{
"error": "Validation error",
"details": "Missing required field: text"
}{
"error": "TTS service error",
"status": 400,
"details": "Upstream provider error: 400 Bad Request - Invalid voice"
}{
"error": "<string>",
"message": "<string>",
"statusCode": 123
}{
"error": "Service error",
"details": "Internal processing error"
}{
"error": "TTS service error",
"status": 400,
"details": "Upstream provider error: 400 Bad Request - Invalid voice"
}Authorizations
API key issued by SLNG. Pass as Authorization: Bearer <token>.
Path Parameters
Target TTS model. Any WebSocket-capable TTS model is supported.
Available options:
cartesia/sonic:3, cartesia/sonic:3.5, deepgram/aura:2, fish/tts:drama-3, fish/tts:s2-pro, fish/tts:s2.1-pro, gradium/tts:default, kugelaudio/kugel:2, murf/murftts:falcon, sarvam/bulbul:v3, slng/deepgram/aura:2-en, slng/fish/tts:s2.1-pro, soniox/tts-rt:v1 Body
application/json
Unmute TTS request. The gateway maps voice, text, and language to the target model's field names and forwards any other field to that model's HTTP API unchanged.
Text to synthesize.
Voice identifier for the selected model.
Optional language hint; models with automatic or voice-selected language ignore it.
Optional pronunciation rewrite, applied where the selected model supports it. mode is always rewrite; supply exactly one of name (a dictionary name, 1–128 characters) or dictionary_id.
Show child attributes
Show child attributes
Response
Synthesis successful.
Binary audio data.