MCPcopy Create free account

hub / github.com/QwenLM/Qwen3-TTS / functions

Functions375 in github.com/QwenLM/Qwen3-TTS

⨍ Functions 375◇ Types & classes 111↳ Endpoints 1
Method__init__
( self, dimension: int = 128, input_dimension: Optional[int] = None, output_di
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:731
Method__init__
( self, *, n_q: int = 8, n_q_semantic: int = 1, **kwargs, )
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:789
Method__init__
(self, config: Qwen3TTSTokenizerV2DecoderConfig)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:825
Method__init__
(self, config: MimiConfig)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:900
Method__init__
(self, config: Qwen3TTSTokenizerV2Config)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:929
Method__init__
(self, in_channels, se_channels, out_channels)
qwen_tts/core/models/modeling_qwen3_tts.py:130
Method__init__
(self, channels, attention_channels=128)
qwen_tts/core/models/modeling_qwen3_tts.py:164
Method__init__
( self, in_channels, out_channels, kernel_size, dilation, )
qwen_tts/core/models/modeling_qwen3_tts.py:248
Method__init__
( self, in_channels, out_channels, res2net_scale=8, se_channels=128,
qwen_tts/core/models/modeling_qwen3_tts.py:274
Method__init__
(self, config: Qwen3TTSSpeakerEncoderConfig)
qwen_tts/core/models/modeling_qwen3_tts.py:318
Method__init__
(self, config: Qwen3TTSTalkerConfig, device=None)
qwen_tts/core/models/modeling_qwen3_tts.py:527
Method__init__
(self, config: Qwen3TTSConfig, device=None)
qwen_tts/core/models/modeling_qwen3_tts.py:562
Method__init__
Qwen3TTSRMSNorm is equivalent to T5LayerNorm
qwen_tts/core/models/modeling_qwen3_tts.py:597
Method__init__
(self, config, layer_idx)
qwen_tts/core/models/modeling_qwen3_tts.py:730
Method__init__
(self, input_size: int, intermediate_size: int, output_size: int, act: str, bias=False)
qwen_tts/core/models/modeling_qwen3_tts.py:809
Method__init__
(self, config, intermediate_size=None)
qwen_tts/core/models/modeling_qwen3_tts.py:843
Method__init__
(self, config: Qwen3TTSConfig, layer_idx: int)
qwen_tts/core/models/modeling_qwen3_tts.py:888
Method__init__
(self, config: Qwen3TTSConfig, layer_idx: int)
qwen_tts/core/models/modeling_qwen3_tts.py:962
Method__init__
(self, config: Qwen3TTSTalkerCodePredictorConfig, embedding_dim: int)
qwen_tts/core/models/modeling_qwen3_tts.py:1019
Method__init__
(self, config: Qwen3TTSTalkerCodePredictorConfig, talker_config: Qwen3TTSTalkerConfig)
qwen_tts/core/models/modeling_qwen3_tts.py:1163
Method__init__
(self, config, layer_idx)
qwen_tts/core/models/modeling_qwen3_tts.py:1349
Method__init__
(self, config)
qwen_tts/core/models/modeling_qwen3_tts.py:1431
Method__init__
(self, config: Qwen3TTSTalkerConfig)
qwen_tts/core/models/modeling_qwen3_tts.py:1571
Method__init__
(self, config: Qwen3TTSConfig)
qwen_tts/core/models/modeling_qwen3_tts.py:1816
Method__init__
( self, tokenizer=None, chat_template=None )
qwen_tts/core/models/processing_qwen3_tts.py:41
Method__init__
( self, mel_dim=128, enc_dim=1024, enc_channels=[512, 512, 512, 512, 1536],
qwen_tts/core/models/configuration_qwen3_tts.py:47
Method__init__
( self, vocab_size=2048, hidden_size=1024, intermediate_size=3072, num
qwen_tts/core/models/configuration_qwen3_tts.py:187
Method__init__
( self, code_predictor_config=None, vocab_size=3072, hidden_size=1024,
qwen_tts/core/models/configuration_qwen3_tts.py:370
Method__init__
(self, model: Qwen3TTSForConditionalGeneration, processor, generate_defaults: Optional[Dict[str, Any]] = None)
qwen_tts/inference/qwen3_tts_model.py:70
Method__init__
(self)
qwen_tts/inference/qwen3_tts_tokenizer.py:57
Method__len__
(self)
finetuning/dataset.py:40
Function_case1b
()
examples/test_model_12hz_base.py:108
Function_case2b
()
examples/test_model_12hz_base.py:140
Function_case3b
()
examples/test_model_12hz_base.py:172
Method_conv_forward
( self, x: Tensor, weight: Tensor, bias: Optional[Tensor] )
qwen_tts/core/tokenizer_25hz/vq/whisper_encoder.py:148
Method_init_weights
(self, module)
qwen_tts/core/models/modeling_qwen3_tts.py:479
Method_init_weights
(self, module)
qwen_tts/core/models/modeling_qwen3_tts.py:512
Function_maybe
(v)
qwen_tts/cli/demo.py:58
Method_to_tensor
(x, dtype=None)
qwen_tts/inference/qwen3_tts_tokenizer.py:288
Method_update_model_kwargs_for_generation
(self, outputs, model_kwargs, is_encoder_decoder=False, num_new_tokens=1)
qwen_tts/core/models/modeling_qwen3_tts.py:1314
Methodapply_chat_template
(self, conversations, chat_template=None, **kwargs)
qwen_tts/core/models/processing_qwen3_tts.py:91
Methodbatch_decode
This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring o
qwen_tts/core/models/processing_qwen3_tts.py:77
Methodcodebook
(self)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:297
Methodcollate_fn
(self, batch)
finetuning/dataset.py:146
Methoddecode
Decodes the given frames into an output audio waveform. Note that the output might be a bit bigger than the input. In that case, any
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:1487
Methoddecode
(self, embed_ind, buffers)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:306
Methoddecode
(self, q_indices: torch.Tensor)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:463
Methoddecode
(self, q_indices: torch.Tensor)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:521
Methoddecode
(self, codes: torch.Tensor)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:707
Methoddecode
(self, codes: torch.Tensor)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:721
Methoddecode
(self, codes: torch.Tensor)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:773
Methoddecode
Decode the given codes to the quantized representation.
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:815
Methoddecode
Decodes the given frames into an output audio waveform. Note that the output might be a bit bigger than the input. In that case, any
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:993
Methoddecode
This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this
qwen_tts/core/models/processing_qwen3_tts.py:84
Functioneager_attention_forward
( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:121
Functioneager_attention_forward
( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio
qwen_tts/core/models/modeling_qwen3_tts.py:634
Methodencode
Encodes the input audio waveform into discrete codes. Args: input_values (`torch.Tensor` of shape `(batch_size, sequence
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:1444
Methodencode
(self, x, buffers)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:300
Methodencode
(self, x: torch.Tensor, n_q: tp.Optional[int] = None)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:441
Methodencode
(self, x: torch.Tensor, n_q: tp.Optional[int] = None)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:517
Methodencode
Encodes the input audio waveform into discrete codes. Args: input_values (`torch.Tensor` of shape `(batch_size, sequence
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:961
Methodextra_repr
(self)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:390
Methodextra_repr
(self)
qwen_tts/core/models/modeling_qwen3_tts.py:612
Methodforward
(self, x)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:108
Methodforward
(self, hidden_states: torch.Tensor)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:143
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:167
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:202
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:266
Methodforward
(self, hidden_state)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:331
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:403
Methodforward
( self, hidden_states: torch.Tensor, speaker_embedding: torch.Tensor, conditio
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:435
Methodforward
(self, code, drop_code=False)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:466
Methodforward
(self, hidden_states, emb=None)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:486
Methodforward
(self, hidden_states, emb)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:505
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:528
Methodforward
( self, hidden_states, # noised input x position_embeddings=None, # rotary position
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:587
Methodforward
(self, hidden_states, scale=1000)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:639
Methodforward
(self, timestep)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:655
Methodforward
( self, hidden_states, timestep, position_embeddings=None, block_diff=None )
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:674
Methodforward
Forward pass of the function. Applies the function to the input elementwise. SnakeBeta ∶= x + 1/b * sin^2 (xa)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:722
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:798
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:828
Methodforward
(self, hidden_states)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:851
Methodforward
(self, x)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:864
Methodforward
(self, x)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:982
Methodforward
(self, mel_spectrogram)
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:1052
Methodforward
( self, hidden_states, condition_vector, speaker_embedding, quantized_
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:1114
Methodforward
Generates a waveform from input code and conditioning parameters.
qwen_tts/core/tokenizer_25hz/modeling_qwen3_tts_tokenizer_v1.py:1256
Methodforward
(self, audio: torch.Tensor, **kwargs)
qwen_tts/core/tokenizer_25hz/vq/speech_vq.py:87
Methodforward
x : torch.Tensor, shape = (n_mels, n_ctx) the mel spectrogram of the audio
qwen_tts/core/tokenizer_25hz/vq/speech_vq.py:278
Methodforward
(self, x: Tensor)
qwen_tts/core/tokenizer_25hz/vq/whisper_encoder.py:157
Methodforward
( self, x: Tensor, cu_seqlens = None, )
qwen_tts/core/tokenizer_25hz/vq/whisper_encoder.py:172
Methodforward
( self, x: Tensor, cu_seqlens = None )
qwen_tts/core/tokenizer_25hz/vq/whisper_encoder.py:278
Methodforward
x : torch.Tensor, shape = (n_mels, n_ctx) the mel spectrogram of the audio
qwen_tts/core/tokenizer_25hz/vq/whisper_encoder.py:338
Methodforward
(self, x, buffers)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:218
Methodforward
(self, x, buffers)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:312
Methodforward
(self, x, n_q: tp.Optional[int] = None)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:388
Methodforward
(self, x, n_q: tp.Optional[int] = None)
qwen_tts/core/tokenizer_25hz/vq/core_vq.py:502
Methodforward
(self, hidden_state)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:189
Methodforward
(self, hidden_state)
qwen_tts/core/tokenizer_12hz/modeling_qwen3_tts_tokenizer_v2.py:204
← previousnext →201–300 of 375, ranked by callers