Apply suggestions from code review

Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
This commit is contained in:
Amir Hussein 2025-09-13 11:27:53 -04:00 committed by GitHub
parent 229883e828
commit 46eb129b5e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 3 additions and 8 deletions

View File

@ -61,9 +61,7 @@ https://tensorboard.dev/experiment/yLE399ZPTzePG8B39jRyOw/
| Decoding method | dev WER | test WER | comment |
|------------------------------------|------------|------------|------------------------------------------|
| modified beam search | 47.6 | 51.2 | --epoch 20, --avg 10 |
| modified beam search | 40.8 | 44.4 | --epoch 20, --avg 10 |
To reproduce the above result, use the following commands for training:

View File

@ -78,16 +78,13 @@ def get_args():
def compute_fbank_gpu(args):
src_dir = Path("data/manifests")
output_dir = Path("data/fbank")
output_dir.mkdir(parents=True, exist_ok=True)
num_jobs = os.cpu_count()
num_mel_bins = 80
sampling_rate = 16000
sr = 16000
dataset_parts = (
"train",
"test1",
"dev",
)
dataset_parts = ("dev", "test1") if args.test else ("train", "test1", "dev")
manifests = read_manifests_if_cached(
prefix="iwslt-ta", dataset_parts=dataset_parts, output_dir=src_dir
)