alkove.VoiceprintLearner

class VoiceprintLearner(
    models: Models,
    *,
    voiceprint: Optional[Voiceprint] = None,
    source: str = 'raw',
)

A print learned from the user’s speech, as the inputs of the web SDK learn it: a window joins only if it sounds like the others.

Methods

offer()

offer(pcm: Any) -> dict

Speech of any length, cut into 1.5 s windows, quiet ones skipped: {accepted, rejected, restarted, progress, ready}.

forget()

forget() -> None

Drops every window: learning starts again from nothing.

Attributes

Attribute	Type	Description
progress	float	0 to 1: the windows accepted so far, out of the 4 (6 s of voice) a print needs.
ready	bool	The print has its 4 windows: Focus can use it.
voiceprint	Optional[Voiceprint]	The print learned so far once ready, else None.