fave_audio_textgrid
fave_audio_textgrid(audio_path, textgrid_path, speakers=0, include_overlaps=True, no_optimize=False, recode_rules=None, labelset_parser=None, point_heuristic=None, vowel_place_config=None, f1_cutoff=np.inf, f2_cutoff=np.inf, ft_config='default', reference_values=ReferenceValues(), fave_aligned=False)
Process a single audio/textgrid pair.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
audio_path |
str | Path | Path to an audio file | required |
textgrid_path |
str | Path | Path to a textgrid | required |
speakers |
int | list[int] | str | Path | optional |
Which speaker(s) to produce data for. Can be a numeric index, or a path to a speaker file, or “all” | 0 |
include_overlaps |
bool | Whether or not to include vowels that are overlapped with speech from other tiers. Defaults to True. |
True |
recode_rules |
str | None | Either a string naming built-in set of recode rules, or path to a custom ruleset. Defaults to None. | None |
labelset_parser |
str | None | Either a string naming a built-in labelset parser, or a path to a custom parser definition. Defaults to None. | None |
point_heuristic |
str | None | Either a string naming a built in point heuristic, or a path to a custom heuristic definition. Defaults to None. | None |
vowel_place_config |
str | None) | A path to a vowel place config file. defaults to None. | None |
f1_cutoff |
float | np.float64 | The maximum considerable F1 value | np.inf |
f2_cutoff |
float | np.float64 | The maximum considerable F2 value | np.inf |
ft_config |
str | None | Either a string naming a built-in fasttrack config file, or a path to a custom config file. Defaults to “default”. | 'default' |
reference_values |
ReferenceValues |
A ReferenceValues object defining a path to a reference corpus Defaults to ReferenceValues() | ReferenceValues() |
fave_aligned |
bool | Were the textgrids generated by classic FAVE align? Defaults to False. | False |
Returns
| Type | Description |
|---|---|
| SpeakerCollection | A new_fave.SpeakerCollection |