icefall/egs/gigaspeech/ASR/local/compute_fbank_gigaspeech.py

96 lines
2.8 KiB
Python
Executable File

#!/usr/bin/env python3
# Copyright 2021 Johns Hopkins University (Piotr Żelasko)
# Copyright 2021 Xiaomi Corp. (Fangjun Kuang)
#
# See ../../../../LICENSE for clarification regarding multiple authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from pathlib import Path
import torch
from lhotse import CutSet, KaldifeatFbank, KaldifeatFbankConfig
# Torch's multithreaded behavior needs to be disabled or
# it wastes a lot of CPU and slow things down.
# Do this outside of main() in case it needs to take effect
# even when we are not invoking the main (e.g. when spawning subprocesses).
torch.set_num_threads(1)
torch.set_num_interop_threads(1)
def compute_fbank_gigaspeech():
in_out_dir = Path("data/fbank")
# number of workers in dataloader
num_workers = 20
# number of seconds in a batch
batch_duration = 1000
subsets = (
"DEV",
"TEST",
# "L",
# "M",
# "S",
# "XS",
)
device = torch.device("cpu")
if torch.cuda.is_available():
device = torch.device("cuda", 0)
extractor = KaldifeatFbank(KaldifeatFbankConfig(device=device))
logging.info(f"device: {device}")
for partition in subsets:
cuts_path = in_out_dir / f"gigaspeech_cuts_{partition}.jsonl.gz"
if cuts_path.is_file():
logging.info(f"{cuts_path} exists - skipping")
continue
raw_cuts_path = in_out_dir / f"gigaspeech_cuts_{partition}_raw.jsonl.gz"
logging.info(f"Loading {raw_cuts_path}")
cut_set = CutSet.from_file(raw_cuts_path)
logging.info("Computing features")
cut_set = cut_set.compute_and_store_features_batch(
extractor=extractor,
storage_path=f"{in_out_dir}/gigaspeech_feats_{partition}",
num_workers=num_workers,
batch_duration=batch_duration,
overwrite=True,
)
cut_set = cut_set.trim_to_supervisions(
keep_overlapping=False, min_duration=None
)
logging.info(f"Saving to {cuts_path}")
cut_set.to_file(cuts_path)
logging.info(f"Saved to {cuts_path}")
def main():
formatter = "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
logging.basicConfig(format=formatter, level=logging.INFO)
compute_fbank_gigaspeech()
if __name__ == "__main__":
main()