Skip to main content
Discover audio endpoints and drive transcription and speech through MCP tools or the Python API.

​
Discover Audio Endpoints

List every registered audio endpoint with the audio tag filter. 
praisonai endpoints list --tags audio

​
MCP Tools

The audio capability is surfaced as two MCP tools. 
praisonai.audio.transcribe(file_path, model="whisper-1", language=None) -> str
praisonai.audio.speech(text, model="tts-1", voice="alloy") -> str
praisonai.audio.transcribe returns the transcript text. praisonai.audio.speech converts text to audio.

​
No Dedicated CLI Verb

There is no top-level praisonai audio or praisonai transcribe subcommand. Use the Python API or the MCP tools above. 
from praisonai.capabilities import transcribe, speech

transcript = transcribe("./meeting.mp3")
print(transcript.text)

audio = speech("Hello, world!", voice="nova")
audio.save("hello.mp3")

​
Environment Variables

Set the API key for whichever provider your model targets.
VariableProvider
OPENAI_API_KEYOpenAI Whisper / TTS
DEEPGRAM_API_KEYDeepgram
ELEVENLABS_API_KEYElevenLabs
export OPENAI_API_KEY=...
praisonai endpoints list --tags audio

Audio (Python)

Full Python usage for transcribe and speech

Endpoints CLI

Explore all registered endpoints