ttslearn.dnntts.gen.gen_waveform

ttslearn.dnntts.gen.gen_waveform(sample_rate, acoustic_features, stream_sizes, has_dynamic_features, num_windows=3, post_filter=False)[source]

Generate waveform from acoustic features.

Parameters
  • sample_rate (int) – sampling rate

  • acoustic_features (numpy.ndarray) – acoustic features

  • stream_sizes (list) – list of stream sizes

  • has_dynamic_features (list) – whether the acoustic features contains dynamic features

  • num_windows (int) – number of windows

  • post_filter (bool) – whether to use post filter

Returns

waveform

Return type

numpy.ndarray