Skip to main content
Version: Next

Function: parseVoice()

parseVoice(base64): Float32Array

Defined in: extensions/speech/utils/kokoroUtils.ts:66

Parses a base64-encoded Kokoro voice file into its raw float rows. Each row holds a KOKORO_VOICE_REF_SIZE-long reference vector for one input token count.

Parameters

base64

string

The base64 contents of the voice .bin file.

Returns

Float32Array

The flattened voice matrix, row-major.

We are Software Mansion.