Skip to main content

Usage guide

Setup, environment variables, and Python/JavaScript/Go usage examples.

​
GroqTTS

Text-to-speech on Groq’s fast inference using Orpheus voice models. Good for very low-latency English and Arabic speech. Authenticates with GROQ_API_KEY from your environment, or an explicit api_key.

​
Fields

​
str
default:"UNSET"
Speech model to synthesize with, e.g. “canopylabs/orpheus-v1-english” or “canopylabs/orpheus-arabic-saudi” for Arabic. Defaults to ‘canopylabs/orpheus-v1-english’.
​
str
default:"UNSET"
Voice to speak with, e.g. “hannah”, “troy” or “austin” for the Orpheus English model. Defaults to ‘hannah’.
​
float
default:"UNSET"
Speaking rate multiplier, 0.5 to 5.0. Defaults to 1.0.
​
Literal['flac', 'mp3', 'mulaw', 'ogg', 'wav']
default:"UNSET"
Audio format returned by Groq: “flac”, “mp3”, “mulaw”, “ogg” or “wav”. Defaults to ‘wav’.
​
int
default:"UNSET"
Output audio sample rate in Hz: 8000, 16000, 22050, 24000, 32000, 44100 or 48000. Defaults to 24000.