Usage: fave-extract audio-textgrid [OPTIONS] AUDIO_PATH
TEXTGRID_PATH
Aliases: audio-textgrid
Run fave-extract on a single audio+textgrid pair.
Positional arguments:
AUDIO_PATH Path to the audio file.
TEXTGRID_PATH Path to the TextGrid file.
Configuration Options:
--recode-rules TEXT Recoding rules to adjust vowel interval
labels. Values can be a string naming one of
the built-in recode rules
('cmu2labov','cmu2phila', 'norecode'), or a
path to a custom recoding yml file.
[default: cmu2labov]
--labelset-parser TEXT A labeleset parser. Values can be a string
naming a built-in parser ('cmu_parser') or a
path to a custom parser yml file. [default:
cmu_parser]
--point-heuristic TEXT The point measurement heuristic to use.
Values can be a built in heuristic ('fave')
or a path to a custom heuristic file.
[default: fave]
--vowel-place TEXT A vowel place definition file. Values can be
the name of a built in config ('defailt) or a
path to a custom config file. [default:
default]
--f1-cutoff FLOAT The maximum considerable F1 value. [default:
1500]
--f2-cutoff FLOAT The maximum considerable F2 value. [default:
3500]
--ft-config TEXT A fasttrack config file. Values can be the
name of a built in config ('default') or a
path to a custom config file. [default:
default]
--fave-aligned Include this flag if the textgrid was aligned
with FAVE align.
--exclude-overlaps Include this flag if you want to exclude
overlapping speech.
--no-optimize Include this flag if you want to skip fave
optimization
Reference Values: [mutually exclusive]
--logparam-reference DIRECTORY A path to a collection of reference
*_logparam.csv files.
--param-reference DIRECTORY A path to a collection of reference
*_param.csv files.
--points-reference DIRECTORY A path to a collection of reference
*_points.csv files.
Output options:
Options for writing output data.
--destination DIRECTORY Destination directory for resulting data
files. If the directory doesn't exist, it
will be created. [default: fave_results]
--which [all|tracks|points|param|log_param|textgrid]
Which output files to write. Default is
'all'. This option can be included multiple
times to write just some of the options (e.g.
'tracks' and 'points' [default: all]
--separate Should each individual speaker be written to
separate data files?
Other options:
--speakers TEXT Which speakers to analyze. Values can be: a
numeric value (1 = first speaker), the string
'all', for all speakers, or a path to a
speaker demographics file. [default: 1]
--help Show this message and exit.