mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-26 18:24:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

minor fixes

Browse Source
This commit is contained in:
jinzr 2024-03-08 02:10:57 +08:00
parent 4e90233785
commit ff53cb0458
3 changed files with 18 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
egs/commonvoice/ASR/pruned_transducer_stateless7/asr_datamodule.py
View File

@ -381,9 +381,11 @@ class CommonVoiceAsrDataModule:
def test_dataloaders(self, cuts: CutSet) -> DataLoader: def test_dataloaders(self, cuts: CutSet) -> DataLoader:
logging.debug("About to create test dataset") logging.debug("About to create test dataset")
test = K2SpeechRecognitionDataset( test = K2SpeechRecognitionDataset(
input_strategy=OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80))) input_strategy=(
if self.args.on_the_fly_feats OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80)))
else eval(self.args.input_strategy)(), if self.args.on_the_fly_feats
else eval(self.args.input_strategy)()
),
return_cuts=self.args.return_cuts, return_cuts=self.args.return_cuts,
) )
sampler = DynamicBucketingSampler( sampler = DynamicBucketingSampler(

14
egs/commonvoice/ASR/pruned_transducer_stateless7_streaming/commonvoice_fr.py
View File

@ -31,7 +31,7 @@ from lhotse.dataset import ( # noqa F401 for PrecomputedFeatures
DynamicBucketingSampler, DynamicBucketingSampler,
K2SpeechRecognitionDataset, K2SpeechRecognitionDataset,
PrecomputedFeatures, PrecomputedFeatures,
SingleCutSampler, SimpleCutSampler,
SpecAugment, SpecAugment,
) )
from lhotse.dataset.input_strategies import ( # noqa F401 For AudioSamples from lhotse.dataset.input_strategies import ( # noqa F401 For AudioSamples
@ -315,8 +315,8 @@ class CommonVoiceAsrDataModule:
drop_last=self.args.drop_last, drop_last=self.args.drop_last,
) )
else: else:
logging.info("Using SingleCutSampler.") logging.info("Using SimpleCutSampler.")
train_sampler = SingleCutSampler( train_sampler = SimpleCutSampler(
cuts_train, cuts_train,
max_duration=self.args.max_duration, max_duration=self.args.max_duration,
shuffle=self.args.shuffle, shuffle=self.args.shuffle,
@ -383,9 +383,11 @@ class CommonVoiceAsrDataModule:
def test_dataloaders(self, cuts: CutSet) -> DataLoader: def test_dataloaders(self, cuts: CutSet) -> DataLoader:
logging.debug("About to create test dataset") logging.debug("About to create test dataset")
test = K2SpeechRecognitionDataset( test = K2SpeechRecognitionDataset(
input_strategy=OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80))) input_strategy=(
if self.args.on_the_fly_feats OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80)))
else eval(self.args.input_strategy)(), if self.args.on_the_fly_feats
else eval(self.args.input_strategy)()
),
return_cuts=self.args.return_cuts, return_cuts=self.args.return_cuts,
) )
sampler = DynamicBucketingSampler( sampler = DynamicBucketingSampler(

8
egs/libriheavy/ASR/zipformer_prompt_asr/asr_datamodule.py
View File

@ -425,9 +425,11 @@ class LibriHeavyAsrDataModule:
def test_dataloaders(self, cuts: CutSet) -> DataLoader: def test_dataloaders(self, cuts: CutSet) -> DataLoader:
logging.debug("About to create test dataset") logging.debug("About to create test dataset")
test = K2SpeechRecognitionDataset( test = K2SpeechRecognitionDataset(
input_strategy=OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80))) input_strategy=(
if self.args.on_the_fly_feats OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80)))
else PrecomputedFeatures(), if self.args.on_the_fly_feats
else PrecomputedFeatures()
),
return_cuts=self.args.return_cuts, return_cuts=self.args.return_cuts,
) )
sampler = DynamicBucketingSampler( sampler = DynamicBucketingSampler(