For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python class

AudioExecutorOutputs

AudioExecutorOutputs

class max.pipelines.audio.AudioExecutorOutputs(waveform)

source

Bases: TensorStruct

One batch of generated waveforms.

Parameters:

waveform (Tensor)

waveform

waveform: Tensor

source

(batch, channels, samples) in [-1, 1]. An audio generator may produce fewer samples than the request asked for, so this is what says how long the audio actually is.