utils#
- diffsptk.get_alpha(sr, mode='hts', n_freq=10, n_alpha=100)[source]#
Compute an appropriate frequency warping factor under given sample rate.
- Parameters:
- srint >= 1
Sample rate in Hz.
- mode[‘hts’, ‘auto’]
‘hts’ returns traditional alpha used in HTS. ‘auto’ computes appropriate alpha in L2 sense.
- n_freqint >= 2
Number of sample points in the frequency domain.
- n_alphaint >= 1
Number of sample points to search alpha.
- Returns:
- outfloat in [0, 1)
Frequency warping factor, \(\alpha\).
Examples
>>> _, sr = diffsptk.read("assets/data.wav") >>> alpha = diffsptk.get_alpha(sr) >>> alpha 0.42
- diffsptk.read(filename, double=False, device=None, **kwargs)[source]#
Read waveform from file.
- Parameters:
- filenamestr
Path of wav file.
- doublebool
If True, return double-type tensor.
- devicetorch.device or None
Device of returned tensor.
- **kwargsadditional keyword arguments
Additional arguments passed to soundfile.read.
- Returns:
- xTensor
Waveform.
- srint
Sample rate in Hz.
Examples
>>> x, sr = diffsptk.read("assets/data.wav") >>> x tensor([ 0.0002, 0.0004, 0.0006, ..., 0.0006, -0.0006, -0.0007]) >>> sr 16000
- diffsptk.write(filename, x, sr, **kwargs)[source]#
Write waveform to file.
- Parameters:
- filenamestr
Path of wav file.
- xTensor
Waveform.
- srint
Sample rate in Hz.
- **kwargsadditional keyword arguments
Additional arguments passed to soundfile.write.
Examples
>>> x, sr = diffsptk.read("assets/data.wav") >>> diffsptk.write("out.wav", x, sr)
- diffsptk.two_pi#
The value of \(2\pi\).