File size: 1,195 Bytes
0b32ad6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 |
from typing import List
from .base import SequentialDataPipe
from .common_pipes import LoadAudio, RandomCrop, SetOutputKeys
class SpeakerVerificationPipe(SequentialDataPipe):
"""
each item in the input dataset should have:
wav_path: str
label: str
"""
def __init__(
self,
audio_sample_rate: int = 16000,
audio_channel_reduction: str = "first",
random_crop_secs: float = -1,
sox_effects: List[List] = None,
):
pipes = [
LoadAudio(
audio_sample_rate=audio_sample_rate,
audio_channel_reduction=audio_channel_reduction,
sox_effects=sox_effects,
),
]
output_keys = dict(
x="wav",
x_len="wav_len",
label="label",
unique_name="id",
)
if random_crop_secs != -1:
pipes.append(
RandomCrop(sample_rate=audio_sample_rate, max_secs=random_crop_secs)
)
output_keys["x"] = "wav_crop"
output_keys["x_len"] = "wav_crop_len"
pipes.append(SetOutputKeys(output_keys))
super().__init__(*pipes)
|