ldc_bpcsad.io.write_htk_label_file

ldc_bpcsad.io.write_htk_label_file(fpath, segs, rec_dur=None, is_sorted=False, in_sec=True, precision=2)

Write speech segments to HTK label file.

The resulting file will contain alternating speech/non-speech segments.

Parameters:

• fpath (pathlib.Path) – Path to file in HTK label file format.

• segs (Iterable[Segment]) – Speech segments.

• rec_dur (float, optional) – Recording duration in seconds. Used to set boundary of final non-speech segment. If None, set to segs[-1].offset. (Default: None)

• is_sorted (bool, optional) – If True, treat segs as already sorted. Otherwise, sort before writing. (Default: False)

• in_sec (bool, optional) – If True, write onsets/offsets in seconds. Else, write onsets/offsets in HTK 100 ns units. (Default: True)

• precision (int, optional) – Output will be truncated to precision decimal places. (Default: 2)

References

[1]

Young, S., Evermann, G., Gales, M., Hain, T., Kershaw, D., Liu, X., … & Woodland, P. (2002). The HTK book. Cambridge University Engineering Department. [link]