alkove.Processor

class Processor(
    models: Models,
    *,
    focus: bool = False,
    voiceprint: Any = None,
    strength: float = 1.0,
    cue: Optional[bool] = None,
    speech: Optional[SpeechParams] = None,
    normalize: bool = False,
    trace: bool = False,
    rate: int = RATE,
    words: Optional[Iterable] = None,
    context: str = 'anywhere',
)

Focus, Cue and Spot over audio pushed in blocks of any size, as real time feeds them: the output comes back aligned with the input, at 16 kHz.

Methods

push()

push(pcm: Any) -> Part

Float32 samples (or int16) at the rate given, any length.

end()

end() -> Part

The samples still inside the models, and the speech still open, closed at the end.