Usage guide
Setup, environment variables, and Python/JavaScript/Go usage examples.
MurfAITTS
Text-to-speech with Murf.ai voices, streamed over WebSocket for low latency. Choose it for Murf’s catalog of natural, styled voices across many accents. Authenticates with MURFAI_API_KEY from your environment, or an explicitapi_key.
Fields
str
default:"UNSET"
Murf region to connect to, e.g. “GLOBAL”, “US_EAST”, “UK” or “INDIA”. Pick the one closest to your users. Defaults to ‘GLOBAL’.
str
default:"UNSET"
Voice model to synthesize with: “Falcon” for low latency or “GEN2” for higher quality. Defaults to ‘Falcon’.
str
default:"UNSET"
Murf voice ID, e.g. “en-US-natalie”. Defaults to ‘en-US-natalie’.
Mapping[str, Any] | None
default:"UNSET"
Advanced voice controls such as pitch, rate and style. Defaults to None (not set).
bool
default:"UNSET"
Whether to stream over WebSocket for low latency; set to False to use HTTP requests instead. Defaults to True.
int
default:"UNSET"
WebSocket only. Minimum number of characters buffered before audio is generated; lower starts faster, higher gives smoother prosody. Defaults to 3.
int
default:"UNSET"
WebSocket only. Maximum milliseconds to wait for more text before generating audio; 0 starts as soon as min_buffer_size is reached. Defaults to 0.
float
default:"UNSET"
Seconds after which the WebSocket connection is refreshed to stay within Murf’s session limits. Defaults to 300.0.