regression

This commit is contained in:
Yuekai Zhang 2024-01-23 17:53:55 +08:00
parent f66b266aa4
commit 08db3051ad

View File

@ -137,7 +137,7 @@ def compute_fbank_wenetspeech_splits(args):
set_audio_duration_mismatch_tolerance(0.01) # 10ms tolerance set_audio_duration_mismatch_tolerance(0.01) # 10ms tolerance
set_caching_enabled(False) set_caching_enabled(False)
with get_executor() as ex: # Initialize the executor only once. #with get_executor() as ex: # Initialize the executor only once.
for i in range(start, stop): for i in range(start, stop):
idx = f"{i + 1}".zfill(num_digits) idx = f"{i + 1}".zfill(num_digits)
logging.info(f"Processing {idx}/{num_splits}") logging.info(f"Processing {idx}/{num_splits}")
@ -158,21 +158,21 @@ def compute_fbank_wenetspeech_splits(args):
) )
logging.info("Computing features") logging.info("Computing features")
# cut_set = cut_set.compute_and_store_features_batch( cut_set = cut_set.compute_and_store_features_batch(
# extractor=extractor,
# storage_path=f"{output_dir}/feats_{subset}_{idx}",
# num_workers=args.num_workers,
# batch_duration=args.batch_duration,
# storage_type=LilcomChunkyWriter,
# overwrite=True,
# )
cut_set = cut_set.compute_and_store_features(
extractor=extractor, extractor=extractor,
storage_path=f"{output_dir}/feats_{subset}_{idx}", storage_path=f"{output_dir}/feats_{subset}_{idx}",
num_jobs=args.num_workers, num_workers=args.num_workers,
executor=ex, batch_duration=args.batch_duration,
storage_type=LilcomChunkyWriter, storage_type=LilcomChunkyWriter,
overwrite=True,
) )
# cut_set = cut_set.compute_and_store_features(
# extractor=extractor,
# storage_path=f"{output_dir}/feats_{subset}_{idx}",
# num_jobs=args.num_workers,
# executor=ex,
# storage_type=LilcomChunkyWriter,
# )
logging.info(f"Saving to {cuts_path}") logging.info(f"Saving to {cuts_path}")
cut_set.to_file(cuts_path) cut_set.to_file(cuts_path)