From 1235e23fbfbc90d35311158aa2a9121c8278f001 Mon Sep 17 00:00:00 2001 From: Fangjun Kuang Date: Sun, 5 Jun 2022 23:44:58 +0800 Subject: [PATCH] Minor fixes. --- egs/gigaspeech/ASR/conformer_ctc/asr_datamodule.py | 6 ++---- egs/librispeech/ASR/conformer_ctc/train.py | 2 ++ 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/egs/gigaspeech/ASR/conformer_ctc/asr_datamodule.py b/egs/gigaspeech/ASR/conformer_ctc/asr_datamodule.py index ec3f75901..62b43146a 100644 --- a/egs/gigaspeech/ASR/conformer_ctc/asr_datamodule.py +++ b/egs/gigaspeech/ASR/conformer_ctc/asr_datamodule.py @@ -363,7 +363,7 @@ class GigaSpeechAsrDataModule: def dev_cuts(self) -> CutSet: logging.info("About to get dev cuts") cuts_valid = load_manifest_lazy( - self.args.manifest_dir / "gigaspeech_cuts_DEV.jsonl.gz" + self.args.manifest_dir / "cuts_DEV.jsonl.gz" ) if self.args.small_dev: return cuts_valid.subset(first=1000) @@ -373,6 +373,4 @@ class GigaSpeechAsrDataModule: @lru_cache() def test_cuts(self) -> CutSet: logging.info("About to get test cuts") - return load_manifest_lazy( - self.args.manifest_dir / "gigaspeech_cuts_TEST.jsonl.gz" - ) + return load_manifest_lazy(self.args.manifest_dir / "cuts_TEST.jsonl.gz") diff --git a/egs/librispeech/ASR/conformer_ctc/train.py b/egs/librispeech/ASR/conformer_ctc/train.py index 5ac74f932..fc8fc8863 100755 --- a/egs/librispeech/ASR/conformer_ctc/train.py +++ b/egs/librispeech/ASR/conformer_ctc/train.py @@ -700,6 +700,8 @@ def run(rank, world_size, args): # the threshold return 1.0 <= c.duration <= 20.0 + train_cuts = train_cuts.filter(remove_short_and_long_utt) + train_dl = librispeech.train_dataloaders(train_cuts) valid_cuts = librispeech.dev_clean_cuts()