Estimators

pyvoimooo.voicing(in, fs)

Estimate the voicing of an input speech signal.

Parameters

  • in (array<float>) – The input wavform to extract the frames from.

  • fs (int, Hz) – The sampling rate.

Returns

  • tss (float,second) - Time instants of analysis (the center time of each frame)

  • voicings (array<float>,[0.0,1.0]).

pyvoimooo.level(in, fs[, useloudness][, voicedonly])

Estimate the level of the voice in the speech signal.

Parameters

  • in (array<float>) – The input wavform to extract the frames from.

  • fs (int, Hz) – The sampling rate.

  • useloudness (boolean, optional) – Weight the spectrum according to perceived loudness (def. true).

  • voicedonly (boolean, optional) – Compute the level only from voiced segments (def. true). Setting it to false will compute the level from whole signal.

Returns

  • lvl (float,[dB]) - The estimated level.

New in version 0.17.14.

pyvoimooo.normalise(in, fs, rqlvl)

Normalise the given waveform to a given level rqlvl using level() function.

Parameters

  • in (array<float>) – The input waveform to transform.

  • fs (int, Hz) – The sampling rate.

  • rqlvl (float, dB) – The requested voice level.

Returns

  • syn (array<float>) - The normalised waveform.

New in version 0.17.14.