espnet2.train.preprocessor.EnhPreprocessor

source

class espnet2.train.preprocessor.EnhPreprocessor(train: bool, rir_scp: str | None = None, rir_apply_prob: float = 1.0, noise_scp: str | None = None, noise_apply_prob: float = 1.0, noise_db_range: str = '3_10', short_noise_thres: float = 0.5, speech_volume_normalize: float | None = None, speech_name: str = 'speech_mix', speech_ref_name_prefix: str = 'speech_ref', noise_ref_name_prefix: str = 'noise_ref', dereverb_ref_name_prefix: str = 'dereverb_ref', use_reverberant_ref: bool = False, num_spk: int = 1, num_noise_type: int = 1, sample_rate: int = 8000, force_single_channel: bool = False, channel_reordering: bool = False, categories: List | None = None, data_aug_effects: List | None = None, data_aug_num: List[int] = [1, 1], data_aug_prob: float = 0.0, speech_segment: int | None = None, avoid_allzero_segment: bool = True, flexible_numspk: bool = False)

Bases: CommonPreprocessor

Preprocessor for Speech Enhancement (Enh) task.

This class is responsible for processing audio data for speech enhancement tasks. It applies techniques such as room impulse response (RIR) convolution and noise addition to the input audio signals, enabling the training of models to enhance speech quality.

rir_scp

Path to the RIR scp file.

• Type: Optional[str]

rir_apply_prob

Probability of applying RIR.

• Type: float

noise_scp

Path to the noise scp file.

• Type: Optional[str]

noise_apply_prob

Probability of applying noise.

• Type: float

noise_db_range

Range of noise levels in dB.

• Type: str

short_noise_thres

Threshold of short noise.

• Type: float

speech_volume_normalize

Volume normalization factor.

• Type: float

speech_name

Key for speech data in input.

• Type: str

speech_ref_name_prefix

Prefix for reference speech keys.

• Type: str

noise_ref_name_prefix

Prefix for noise reference keys.

• Type: str

dereverb_ref_name_prefix

Prefix for dereverberated reference keys.

• Type: str

use_reverberant_ref

Flag to use reverberant reference.

• Type: bool

num_spk

Number of speakers.

• Type: int

num_noise_type

Number of noise types.

• Type: int

sample_rate

Sampling rate for audio processing.

• Type: int

force_single_channel

Flag to convert to single-channel audio.

• Type: bool

channel_reordering

Flag to randomly reorder channels.

• Type: bool

categories

Mapping of categories to unique integers.

• Type: Optional[List]

data_aug_effects

Data augmentation effects to apply.

• Type: List

data_aug_num

Number of augmentations to apply.

• Type: List[int]

data_aug_prob

Probability of applying data augmentation.

• Type: float

speech_segment

Length of speech segments for processing.

• Type: Optional[int]

avoid_allzero_segment

Flag to avoid all-zero segments.

• Type: bool

flexible_numspk

Flag to allow variable number of speakers.

• Type: bool

• Parameters:

  • train (bool) – Whether to use in training mode.
  • rir_scp (Optional *[*str ]) – Path to the RIR scp file.
  • rir_apply_prob (float) – Probability of applying RIR.
  • noise_scp (Optional *[*str ]) – Path to the noise scp file.
  • noise_apply_prob (float) – Probability of applying noise.
  • noise_db_range (str) – Range of noise levels in dB.
  • short_noise_thres (float) – Threshold of short noise.
  • speech_volume_normalize (float) – Volume normalization factor.
  • speech_name (str) – Key for speech data in input.
  • speech_ref_name_prefix (str) – Prefix for reference speech keys.
  • noise_ref_name_prefix (str) – Prefix for noise reference keys.
  • dereverb_ref_name_prefix (str) – Prefix for dereverberated reference keys.
  • use_reverberant_ref (bool) – Flag to use reverberant reference.
  • num_spk (int) – Number of speakers.
  • num_noise_type (int) – Number of noise types.
  • sample_rate (int) – Sampling rate for audio processing.
  • force_single_channel (bool) – Flag to convert to single-channel audio.
  • channel_reordering (bool) – Flag to randomly reorder channels.
  • categories (Optional *[*List ]) – Mapping of categories to unique integers.
  • data_aug_effects (List) – Data augmentation effects to apply.
  • data_aug_num (List *[*int ]) – Number of augmentations to apply.
  • data_aug_prob (float) – Probability of applying data augmentation.
  • speech_segment (Optional *[*int ]) – Length of speech segments for processing.
  • avoid_allzero_segment (bool) – Flag to avoid all-zero segments.
  • flexible_numspk (bool) – Flag to allow variable number of speakers.

Examples

>>> preprocessor = EnhPreprocessor(train=True, rir_scp="path/to/rir.scp",
...                                 noise_scp="path/to/noise.scp")
>>> processed_data = preprocessor(uid="sample_uid", data={"speech": audio_data})

• Raises:ValueError – If any of the input parameters are invalid or inconsistent.

NOTE

Ensure that the sampling rates of all audio data and RIRs are consistent to avoid processing errors.