def get_data(featurizer: TextFeaturizer):
train_dataset = ASRSliceDataset(data_paths=[data_path],
speech_featurizer=speech_featurizer,
text_featurizer=featurizer,
stage="train",
shuffle=False)
train_data = train_dataset.create(1)
return next(iter(train_data))
test_dataset = ASRSliceDataset(speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer,
**vars(
config.learning_config.test_dataset_config))
# build model
rnn_transducer = RnnTransducer(**config.model_config,
vocabulary_size=text_featurizer.num_classes)
rnn_transducer.make(speech_featurizer.shape)
rnn_transducer.load_weights(args.saved, by_name=True)
rnn_transducer.summary(line_length=100)
rnn_transducer.add_featurizers(speech_featurizer, text_featurizer)
batch_size = args.bs or config.learning_config.running_config.batch_size
test_data_loader = test_dataset.create(batch_size)
with file_util.save_file(file_util.preprocess_paths(args.output)) as filepath:
overwrite = True
if tf.io.gfile.exists(filepath):
overwrite = input(
f"Overwrite existing result file {filepath} ? (y/n): ").lower(
) == "y"
if overwrite:
results = rnn_transducer.predict(test_data_loader, verbose=1)
logger.info(f"Saving result to {args.output} ...")
with open(filepath, "w") as openfile:
openfile.write("PATH\tDURATION\tGROUNDTRUTH\tGREEDY\tBEAMSEARCH\n")
progbar = tqdm(total=test_dataset.total_steps, unit="batch")
for i, pred in enumerate(results):
groundtruth, greedy, beamsearch = [
config = Config(config_dir, learning=True)
speech_featurizer = TFSpeechFeaturizer(config.speech_config)
text_featurizer = CharFeaturizer(config.decoder_config)
train_dataset = ASRSliceDataset(
data_paths=config.learning_config.dataset_config.train_paths,
speech_featurizer=speech_featurizer,
text_featurizer=text_featurizer,
augmentations=config.learning_config.augmentations,
stage="train",
cache=False,
shuffle=True,
sort=False)
train_data = train_dataset.create(2)
train_data_loader = strategy.experimental_distribute_dataset(train_data)
train_iterator = iter(train_data_loader)
while True:
batch = next(train_iterator)
features, input_length, labels, label_length, prediction, prediction_length = batch
#print ("features")
#print (features)
#print ("input_length")
#print (input_length)
print("labels")
print(labels)
#print ("label_length")
#print (label_length)