MCPcopy Create free account

hub / github.com/Z-yq/TensorflowASR / functions

Functions1,065 in github.com/Z-yq/TensorflowASR

⨍ Functions 1,065◇ Types & classes 251↳ Endpoints 6
↓ 176 callersFunctionThrowOnError
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_inline.h:12
↓ 65 callersMethodsize
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_api.h:813
↓ 49 callersMethodget_config
(self)
asr/models/conformer_blocks.py:23
↓ 36 callersMethodget_config
(self)
asr/models/chunk_conformer_blocks.py:17
↓ 31 callersMethodupdate_state
(self, decode: tf.Tensor, target: tf.Tensor)
utils/metrics.py:45
↓ 30 callersMethodget
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_api.h:812
↓ 25 callersMethodsoftmax
(self,logits)
Inference/PythonInference/asr/src/asr.py:27
↓ 19 callersMethodpad
(self, x, mode=1)
punc_recover/dataloaders/punc_dataloader.py:139
↓ 16 callersFunctionnsw_test_case
(raw_text)
augmentations/tts_for_asr/utils/normalize.py:698
↓ 16 callersMethodrun
(self,)
asr/tester/am_tester.py:159
↓ 15 callersMethodstream_call
(self, outputs, mha_caches, cnn_caches)
asr/models/chunk_conformer_blocks.py:750
↓ 13 callersMethodextract
Convert string to a list of integers Args: text: string (sequence of characters) Returns: sequence o
augmentations/tts_for_asr/utils/text_featurizers.py:82
↓ 12 callersMethodcheck_valid
(self, txt, vocab_list)
asr/dataloaders/am_dataloader.py:229
↓ 12 callersMethodload
Inference/CppInference/onnx/ext/AudioFile.hpp:450
↓ 12 callersMethodresult
(self)
utils/metrics.py:50
↓ 11 callersMethod__init__
(self, axis=-1, name="glu_activation", **kwargs)
asr/models/conformer_blocks.py:11
↓ 11 callersMethodcompile
(self,)
asr/tester/am_tester.py:91
↓ 10 callersMethod__init__
(self, axis=-1, name="glu_activation", **kwargs)
asr/models/chunk_conformer_blocks.py:8
↓ 10 callersMethod_build
(self)
asr/models/wav_model.py:48
↓ 10 callersMethodload_wav
(self,path)
utils/speech_featurizers.py:68
↓ 9 callersMethodcompile
(self)
test_asr.py:26
↓ 9 callersMethoddecode
(self,enc_features)
Inference/PythonInference/asr/src/asr.py:63
↓ 9 callersMethodget_shape
(self, shape)
test_chunk_asr.py:159
↓ 9 callersMethodiextract
Args: feat: tf.Tensor with dim [B, None] Returns: transcripts: tf.Tensor of dtype tf.string with dim [B]
utils/text_featurizers.py:85
↓ 9 callersMethodreturn_data_shape
(self)
asr/dataloaders/am_dataloader.py:41
↓ 9 callersMethodreturn_data_types
(self)
asr/dataloaders/am_dataloader.py:37
↓ 9 callersFunctionshape_list
Deal with dynamic shape in tensorflow cleanly.
utils/tools.py:132
↓ 8 callersMethodcheck_valid
(self, txt, vocab_list)
asr/dataloaders/chunk_dataloader.py:252
↓ 8 callersMethodendid
(self)
augmentations/tts_for_asr/utils/text_featurizers.py:78
↓ 8 callersMethodeval_per_epoch_steps
(self)
asr/dataloaders/am_dataloader.py:54
↓ 7 callersMethodAdd
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_inline.h:355
↓ 7 callersFunctionnum2chn
(number_string, numbering_type=NUMBERING_TYPES[1], big=False, traditional=False, alt_zero=False, a
augmentations/tts_for_asr/utils/normalize.py:311
↓ 7 callersMethodset_inference_func
(self)
vad/models/vad_model.py:100
↓ 6 callersMethod__init__
Initializes PANN model. Args: depth: the number of convolutional blocks. First block has 64 filters and the number of f
leaf_audio/models.py:186
↓ 6 callersMethodctc_acc
(self, labels, y_pred)
asr/models/chunk_conformer_blocks.py:1009
↓ 6 callersMethodinference
( inputs)
vad/models/vad_model.py:49
↓ 6 callersMethodset_datasets
(self,evaldataset)
asr/tester/base_tester.py:62
↓ 5 callersMethod__init__
( self, learn_pooling: bool = True, learn_filters: bool = True,
leaf_audio/frontend.py:86
↓ 5 callersMethodavailable
(self)
augmentations/augments.py:193
↓ 5 callersFunctionbytes_to_string
(array: np.ndarray, encoding: str = "utf-8")
utils/tools.py:46
↓ 5 callersMethodcardinal2chntext
(self)
augmentations/tts_for_asr/utils/normalize.py:424
↓ 5 callersMethodextract_feature
(self,wav)
Inference/PythonInference/asr/src/asr.py:34
↓ 5 callersMethodinit_cache
(self, B)
asr/models/chunk_conformer_blocks.py:155
↓ 5 callersMethodinit_caches
(self, B)
asr/models/chunk_conformer_blocks.py:727
↓ 5 callersMethodpad_signal
(self,wavs,max_length)
utils/speech_featurizers.py:75
↓ 5 callersFunctionpreprocess_paths
(paths)
utils/tools.py:36
↓ 4 callersMethodRun
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_inline.h:554
↓ 4 callersMethod__init__
Initialize.
vad/utils/stft.py:130
↓ 4 callersMethod__init__
(self, num_layers, d_model,embedding_dim, num_heads, dff, input_vocab_size, maximum_position_
punc_recover/models/punc_transformer.py:200
↓ 4 callersMethod__str__
(self)
augmentations/tts_for_asr/utils/normalize.py:61
↓ 4 callersFunctionchn2num
(chinese_string, numbering_type=NUMBERING_TYPES[1])
augmentations/tts_for_asr/utils/normalize.py:224
↓ 4 callersFunctioncreate_padding_mask
(seq)
punc_recover/models/punc_transformer.py:26
↓ 4 callersMethodendid
(self)
utils/text_featurizers.py:77
↓ 4 callersMethodfeature_pick
Length regulator logic.
asr/models/chunk_conformer_blocks.py:913
↓ 4 callersMethodfit
(self, epoch=None)
asr/trainer/ctc_runners.py:228
↓ 4 callersFunctionnan_to_zero
(input_tensor)
utils/tools.py:42
↓ 4 callersMethodpredict
(self, x)
asr/models/chunk_conformer_blocks.py:815
↓ 4 callersMethodpunc_recover
(self, txt)
Inference/PythonInference/punc_recover/src/punc_recover.py:46
↓ 4 callersMethodrun
(self, )
vad/tester/vad_tester.py:50
↓ 4 callersMethodrun
(self,)
punc_recover/tester/base_tester.py:60
↓ 4 callersMethodsave
Inference/CppInference/onnx/ext/AudioFile.hpp:797
↓ 4 callersMethodset_all_steps
(self,all_steps)
asr/tester/base_tester.py:47
↓ 4 callersMethodset_progbar
(self,total_steps)
asr/tester/base_tester.py:49
↓ 4 callersFunctionsinc_impulse_response
Computes the sinc impulse response.
leaf_audio/impulse_responses.py:55
↓ 4 callersMethodstreaming_live_out
(self)
Inference/PythonInference/stream_asr_session.py:443
↓ 3 callersMethod__init__
(self)
asr/models/wav_model.py:29
↓ 3 callersMethod__init__
(self, kernel_size, strides=1, padding='same',
leaf_audio/pooling.py:152
↓ 3 callersMethod__init__
(self, date=None, chntext=None)
augmentations/tts_for_asr/utils/normalize.py:500
↓ 3 callersMethod_build
(self)
vad/models/vad_model.py:26
↓ 3 callersMethod_compute_chunk_mask
(self, query, win_front, win_back)
asr/models/chunk_conformer_blocks.py:158
↓ 3 callersMethod_create_norm_layer
(self, name)
asr/models/layers/LayerNormLstmCell.py:91
↓ 3 callersMethod_finished
(self,)
asr/trainer/base_runners.py:131
↓ 3 callersMethod_write_to_tensorboard
Write variables to tensorboard.
asr/trainer/base_runners.py:33
↓ 3 callersMethod_write_to_tensorboard
Write variables to tensorboard.
vad/trainer/base_trainer.py:33
↓ 3 callersMethod_write_to_tensorboard
Write variables to tensorboard.
punc_recover/trainer/base_trainers.py:33
↓ 3 callersMethodadd_chunk_size
(self,chunk_size,mel_size,hop_size)
asr/models/conformer_blocks.py:568
↓ 3 callersMethodcall
Calculate forward propagation. Args: y (Tensor): Groundtruth signal (B, T). x (Tensor): Predicted signal (B, T).
vad/utils/stft.py:139
↓ 3 callersMethodconvert
:param wav:语音数据，[-1,1] :param spk: 0~1882 :return: 音色转换后的语音数据
augmentations/tts_for_asr/vc_aug.py:10
↓ 3 callersMethodcreate
(cls, index, value, numbering_type=NUMBERING_TYPES[1], small_unit=False)
augmentations/tts_for_asr/utils/normalize.py:85
↓ 3 callersMethodendid
(self)
Inference/PythonInference/utils/text_featurizers.py:68
↓ 3 callersMethodgetSampleRate
Inference/CppInference/onnx/ext/AudioFile.hpp:287
↓ 3 callersMethodget_per_epoch_steps
(self)
asr/dataloaders/am_dataloader.py:51
↓ 3 callersFunctionmerge_two_last_dims
(x)
utils/tools.py:89
↓ 3 callersMethodnormalize
(self)
augmentations/tts_for_asr/utils/normalize.py:611
↓ 3 callersMethodonly_chinese
(self, word)
asr/dataloaders/am_dataloader.py:108
↓ 3 callersMethodonly_chinese
(self, word)
asr/dataloaders/chunk_dataloader.py:136
↓ 3 callersMethodphone_call
(self, label_gts, training=False)
asr/models/chunk_conformer_blocks.py:745
↓ 3 callersFunctionpreprocess_paths
(paths)
augmentations/tts_for_asr/utils/text_featurizers.py:9
↓ 3 callersMethodprocess
(self, wav)
augmentations/augments.py:213
↓ 3 callersMethodsave_checkpoint
Save checkpoint.
asr/trainer/ctc_runners.py:272
↓ 3 callersMethodset_total_train_steps
(self,value)
asr/trainer/base_runners.py:81
↓ 3 callersMethodsplit_heads
Split the last dimension into (num_heads, depth). Transpose the result such that the shape is (batch_size, num_heads, seq_len, depth)
punc_recover/models/punc_transformer.py:94
↓ 3 callersFunctionwer
(r,h)
utils/xer.py:211
↓ 2 callersMethodFree
Inference/CppInference/onnx/ext/onnxruntime/include/onnxruntime_cxx_inline.h:108
↓ 2 callersMethod__init__
(self, filters: int, kernel_size: int, strides: int,
leaf_audio/convolution.py:249
↓ 2 callersMethod__init__
PCEN constructor. Args: alpha: float, exponent of EMA smoother smooth_coef: float, smoothing coefficient of EMA
leaf_audio/postprocessing.py:71
↓ 2 callersMethod_build
(self)
punc_recover/models/punc_transformer.py:279
↓ 2 callersMethod_calculate_log_mels_spectrogram
Calculate forward propagation. Args: signals (Tensor): signal (B, T). Returns: Tensor: Mel spectrogram (B, T',
vad/utils/stft.py:43
↓ 2 callersMethod_check_log_interval
Save log interval.
asr/trainer/base_runners.py:209
↓ 2 callersMethod_check_log_interval
Save log interval.
vad/trainer/base_trainer.py:209
next →1–100 of 1,065, ranked by callers