paddlespeech.s2t.frontend.augmentor.resample module

Contain the resample augmentation model.

class paddlespeech.s2t.frontend.augmentor.resample.ResampleAugmentor(rng, new_sample_rate)[source]

Bases: AugmentorBase

Augmentation model for resampling.

See more info here: https://ccrma.stanford.edu/~jos/resample/index.html

Parameters:

  • rng (random.Random) -- Random generator object.

  • new_sample_rate (int) -- New sample rate in Hz.

Methods

__call__(x[, uttid, train])

Call self as a function.

transform_audio(audio_segment)

Resamples the input audio to a target sample rate.

transform_feature(spec_segment)

Adds various effects to the input audo feature segment.
transform_audio(audio_segment)[source]

Resamples the input audio to a target sample rate.

Note that this is an in-place transformation.

Parameters:

audio (AudioSegment|SpeechSegment) -- Audio segment to add effects to.