Real-time audio streaming and transcription.
Real-time WebSocket connection for audio streaming and transcription
Transport method for communication
Unique identifier for the session/context (4-24 characters, alphanumeric)
Identifier for a specific flow
Language code for speech recognition
IANA timezone identifier
Voice activity detection configuration (JSON string)
Speech-to-Text Socket.io API channel for audio streaming and LLM events.
To connect to the Socket.io backend correctly, ensure you use the /event namespace.
Available only on servers:
Accepts the following message:
Raw PCM audio chunk.
must be in 16KHz, 16-bit, mono format.
{}
Speech-to-Text Socket.io API channel for audio streaming and LLM events.
To connect to the Socket.io backend correctly, ensure you use the /event namespace.
Available only on servers:
Accepts the following message:
{
"data": [
"hello",
"world"
]
}
Speech-to-Text Socket.io API channel for audio streaming and LLM events.
To connect to the Socket.io backend correctly, ensure you use the /event namespace.
Available only on servers:
Accepts the following message:
{
"data": {
"name": "Transcript",
"value": {
"data": {
"transcript": "Item one is three degrees",
"itn_transcript": "Item 1 is 3 °"
}
}
}
}
Speech-to-Text Socket.io API channel for audio streaming and LLM events.
To connect to the Socket.io backend correctly, ensure you use the /event namespace.
Available only on servers:
Accepts the following message:
{
"message": "string",
"details": "string"
}
Speech-to-Text Socket.io API channel for audio streaming and LLM events.
To connect to the Socket.io backend correctly, ensure you use the /event namespace.
Available only on servers:
Accepts the following message:
{
"results": {}
}
Raw PCM audio chunk.
must be in 16KHz, 16-bit, mono format.