Module blindai.audio

Classes

class Audio

Static methods

def transcribe(file: Union[str, bytes], model: str = 'tiny.en', connection: Optional[ForwardRef('BlindAiConnection')] = None, tee: Optional[str] = 'sgx') ‑> str

BlindAI Whisper API which converts speech to text based on the model passed.

Args

file

str, bytes Audio file to transcribe. It may also receive serialized bytes of wave file.

model

str The Whisper model. Defaults to "medium".

connection

Optional[BlindAiConnection] The BlindAI connection object. Defaults to None.

tee

Optional[str] The Trusted Execution Environment to use. Defaults to "sgx". Unused, at the moment.

Returns

Dict: The transcription object containing, text and the tokens