pnpl.tasks.libribrain.PhonemeClassification

class pnpl.tasks.libribrain.PhonemeClassification(tmin=0.0, tmax=0.5, label_type='phoneme', exclude_phonemes=<factory>, _phonemes_sorted=<factory>, _phoneme_to_id=<factory>)

Multi-class phoneme classification task.

Each sample is aligned to a phoneme onset in the events file.

Parameters:

• tmin (float) – Start time relative to phoneme onset (seconds)

• tmax (float) – End time relative to phoneme onset (seconds)

• label_type (str) – ‘phoneme’ for multi-class, ‘voicing’ for binary (voiced/unvoiced)

• exclude_phonemes (list) – List of phonemes to exclude

• _phonemes_sorted (list)

• _phoneme_to_id (dict)

__init__(tmin=0.0, tmax=0.5, label_type='phoneme', exclude_phonemes=<factory>, _phonemes_sorted=<factory>, _phoneme_to_id=<factory>)

Parameters:

• tmin (float)

• tmax (float)

• label_type (str)

• exclude_phonemes (list)

• _phonemes_sorted (list)

• _phoneme_to_id (dict)

Return type:

None

Methods

__init__([tmin, tmax, label_type, ...])
collect_samples(dataset)	Collect phoneme samples from all runs.
get_label(sample)	Extract label ID from sample.

Attributes

label_info	Label metadata.
label_type
tmax
tmin
exclude_phonemes