MCPcopy Create free account

hub / github.com/Soul-AILab/SoulX-Podcast / functions

Functions193 in github.com/Soul-AILab/SoulX-Podcast

⨍ Functions 193◇ Types & classes 55↳ Endpoints 6
Method__init__
( self, idim: int, hidden_units: int, dropout_rate: float,
soulxpodcast/models/modules/flow_components/upsample_encoder.py:687
Method__init__
( self, size: int, self_attn: torch.nn.Module, feed_forward: Optional[nn.Modul
soulxpodcast/models/modules/flow_components/upsample_encoder.py:731
Method__init__
( self, input_size: int = 512, output_size: int = 512, attention_heads: int =
soulxpodcast/models/modules/flow_components/upsample_encoder.py:859
Method__init__
(self, in_features, out_features, alpha=1.0, alpha_trainable=True, alpha_logscale=True)
soulxpodcast/models/modules/flow_components/estimator.py:57
Method__init__
( self, dim: int, dim_out: Optional[int] = None, mult: int = 4, dropou
soulxpodcast/models/modules/flow_components/estimator.py:108
Method__init__
( self, dim: int, num_attention_heads: int, attention_head_dim: int, d
soulxpodcast/models/modules/flow_components/estimator.py:171
Method__init__
(self, dim)
soulxpodcast/models/modules/flow_components/estimator.py:332
Method__init__
(self, dim, dim_out, time_emb_dim, groups=8)
soulxpodcast/models/modules/flow_components/estimator.py:364
Method__init__
(self, dim)
soulxpodcast/models/modules/flow_components/estimator.py:382
Method__init__
( self, in_channels: int, time_embed_dim: int, act_fn: str = "silu", o
soulxpodcast/models/modules/flow_components/estimator.py:391
Method__init__
(self, channels, use_conv=False, use_conv_transpose=True, out_channels=None, name="conv")
soulxpodcast/models/modules/flow_components/estimator.py:452
Method__init__
(self, dim0: int, dim1: int)
soulxpodcast/models/modules/flow_components/estimator.py:480
Method__init__
( self, in_channels: int, out_channels: int, kernel_size: int, stride:
soulxpodcast/models/modules/flow_components/estimator.py:491
Method__init__
(self, dim: int, dim_out: int)
soulxpodcast/models/modules/flow_components/estimator.py:520
Method__init__
(self, dim: int, dim_out: int, time_emb_dim: int, groups: int = 8)
soulxpodcast/models/modules/flow_components/estimator.py:536
Method__init__
( self, in_channels, out_channels, channels=(256, 256), dropout=0.05,
soulxpodcast/models/modules/flow_components/estimator.py:559
Method__init__
( self, in_channels=320, out_channels=80, channels=[256], # noqa drop
soulxpodcast/models/modules/flow_components/estimator.py:779
Method__init__
( self, channels: int = 512, kernel_size: int = 3, dilations: List[int] = [1,
soulxpodcast/models/modules/hifigan_components/layers.py:92
Method__init__
(self, samp_rate, harmonic_num=0, sine_amp=0.1, noise_std=0.003, voiced_thre
soulxpodcast/models/modules/hifigan_components/layers.py:169
Method__init__
(self, sampling_rate, upsample_scale, harmonic_num=0, sine_amp=0.1, add_noise_std=0.003, voic
soulxpodcast/models/modules/hifigan_components/layers.py:236
Method__init__
(self, samp_rate, upsample_scale, harmonic_num=0, sine_amp=0.1, noise_std=0.003,
soulxpodcast/models/modules/hifigan_components/layers.py:286
Method__init__
(self, sampling_rate, upsample_scale, harmonic_num=0, sine_amp=0.1, add_noise_std=0.003, voic
soulxpodcast/models/modules/hifigan_components/layers.py:404
Method__init__
(self)
api/tasks.py:53
Method__init__
初始化模型（单例模式，只初始化一次）
api/service.py:39
Method__len__
(self)
soulxpodcast/utils/dataloader.py:79
Method__new__
(cls)
api/tasks.py:47
Method__new__
(cls)
api/service.py:30
Method__post_init__
(self)
soulxpodcast/config.py:138
Method__post_init__
确保目录存在并验证配置
api/config.py:56
Function_change_component_language
(lang)
webui.py:510
Function_ras_sample_hf_engine
r""" Generates sequences of token ids for models with a language modeling head using **multinomial sampling** and can be used for text-decoder
soulxpodcast/models/modules/sampler.py:24
Functioncheck_dialect_prompt_text
(text: str, prefix: str = None)
webui.py:218
Functioncheck_dialect_prompt_text
(text: str, prefix: str = None)
soulxpodcast/utils/text.py:60
Functioncheck_dialogue_text
(text_list: List[str])
soulxpodcast/utils/text.py:71
Functioncheck_models
(model_path, inputs)
soulxpodcast/utils/infer_utils.py:91
Functiondialogue_synthesis_function
( target_text: str, spk1_prompt_text: str | None = "", spk1_prompt_audio: str | None = None, s
webui.py:296
Functiondownload_file
下载生成的音频文件
api/main.py:330
Functiondynamic_range_compression
(x, C=1, clip_val=1e-5)
soulxpodcast/utils/audio.py:14
Functiondynamic_range_decompression
(x, C=1)
soulxpodcast/utils/audio.py:18
Functionformat_audio_duration
格式化音频时长
api/utils.py:170
Methodforward
Forward diffusion Args: mu (torch.Tensor): output of encoder shape: (batch_size, n_feats, mel_timesteps)
soulxpodcast/models/modules/flow.py:41
Methodforward
(self, token, token_len, prompt_feat, prompt_f
soulxpodcast/models/modules/flow.py:157
Methodforward
(self, x: torch.Tensor)
soulxpodcast/models/modules/hifigan.py:69
Methodforward
(self, speech_feat: torch.Tensor, cache_source: torch.Tensor = torch.zeros(1, 1, 0))
soulxpodcast/models/modules/hifigan.py:238
Methodforward
Add positional encoding. Args: x (torch.Tensor): Input tensor (batch, time, `*`). Returns: torch.Tensor: Enc
soulxpodcast/models/modules/flow_components/upsample_encoder.py:202
Methodforward
Input x. Args: x (torch.Tensor): Input tensor (#batch, time, idim). x_mask (torch.Tensor): Input mask (#batch, 1, tim
soulxpodcast/models/modules/flow_components/upsample_encoder.py:272
Methodforward
(self, inputs: torch.Tensor, input_lengths: torch.Tensor)
soulxpodcast/models/modules/flow_components/upsample_encoder.py:322
Methodforward
inputs: (batch_size, seq_len, channels)
soulxpodcast/models/modules/flow_components/upsample_encoder.py:344
Methodforward
Compute scaled dot product attention. Args: query (torch.Tensor): Query tensor (#batch, time1, size). key (torch.Tens
soulxpodcast/models/modules/flow_components/upsample_encoder.py:471
Methodforward
Compute 'Scaled Dot Product Attention' with rel. positional encoding. Args: query (torch.Tensor): Query tensor (#batch, time1, siz
soulxpodcast/models/modules/flow_components/upsample_encoder.py:590
Methodforward
Forward function. Args: xs: input tensor (B, L, D) Returns: output tensor, (B, L, D)
soulxpodcast/models/modules/flow_components/upsample_encoder.py:700
Methodforward
Compute encoded features. Args: x (torch.Tensor): (#batch, time, size) mask (torch.Tensor): Mask tensor for the input
soulxpodcast/models/modules/flow_components/upsample_encoder.py:761
Methodforward
Embed positions in tensor. Args: xs: padded input tensor (B, T, D) xs_lens: input length (B) decoding_chu
soulxpodcast/models/modules/flow_components/upsample_encoder.py:923
Methodforward
Forward pass of the function. Applies the function to the input elementwise. SnakeBeta ∶= x + 1/b * sin^2 (xa)
soulxpodcast/models/modules/flow_components/estimator.py:76
Methodforward
(self, hidden_states)
soulxpodcast/models/modules/flow_components/estimator.py:143
Methodforward
( self, hidden_states: torch.FloatTensor, attention_mask: Optional[torch.FloatTensor]
soulxpodcast/models/modules/flow_components/estimator.py:255
Methodforward
(self, x, scale=1000)
soulxpodcast/models/modules/flow_components/estimator.py:337
Methodforward
(self, x, mask, time_emb)
soulxpodcast/models/modules/flow_components/estimator.py:373
Methodforward
(self, x)
soulxpodcast/models/modules/flow_components/estimator.py:386
Methodforward
(self, sample, condition=None)
soulxpodcast/models/modules/flow_components/estimator.py:423
Methodforward
(self, inputs)
soulxpodcast/models/modules/flow_components/estimator.py:466
Methodforward
(self, x: torch.Tensor)
soulxpodcast/models/modules/flow_components/estimator.py:485
Methodforward
(self, x: torch.Tensor)
soulxpodcast/models/modules/flow_components/estimator.py:513
Methodforward
(self, x: torch.Tensor, mask: torch.Tensor)
soulxpodcast/models/modules/flow_components/estimator.py:530
Methodforward
Forward pass of the UNet1DConditional model. Args: x (torch.Tensor): shape (batch_size, in_channels, time) mask (_typ
soulxpodcast/models/modules/flow_components/estimator.py:676
Methodforward
Forward pass of the UNet1DConditional model. Args: x (torch.Tensor): shape (batch_size, in_channels, time) mask (_typ
soulxpodcast/models/modules/flow_components/estimator.py:885
Methodforward
Forward pass of the function. Applies the function to the input elementwise. Snake ∶= x + 1/a * sin^2 (xa)
soulxpodcast/models/modules/hifigan_components/layers.py:76
Methodforward
(self, x: torch.Tensor)
soulxpodcast/models/modules/hifigan_components/layers.py:138
Methodforward
:param f0: [B, 1, sample_len], Hz :return: [B, 1, sample_len]
soulxpodcast/models/modules/hifigan_components/layers.py:185
Methodforward
Sine_source, noise_source = SourceModuleHnNSF(F0_sampled) F0_sampled (batchsize, length, 1) Sine_source (batchsize, length, 1
soulxpodcast/models/modules/hifigan_components/layers.py:251
Methodforward
sine_tensor, uv = forward(f0) input F0: tensor(batchsize=1, length, dim=1) f0 for unvoiced steps should be 0 output
soulxpodcast/models/modules/hifigan_components/layers.py:358
Methodforward
Sine_source, noise_source = SourceModuleHnNSF(F0_sampled) F0_sampled (batchsize, length, 1) Sine_source (batchsize, length, 1
soulxpodcast/models/modules/hifigan_components/layers.py:419
Methodgenerate
( self, prompt: list[str], sampling_param: SamplingParams, past_key_values=Non
soulxpodcast/engine/llm_engine.py:37
Functiongenerate_async
异步生成语音（返回任务ID） 适用于长音频生成或批量任务
api/main.py:226
Functiongenerate_sync
同步生成语音（直接返回音频文件） 适用于短音频生成（预计<30秒）
api/main.py:142
Functionget_task_status
查询任务状态
api/main.py:304
Functionglobal_exception_handler
全局异常处理
api/main.py:345
Functionhealth_check
健康检查
api/main.py:126
Functioninit_weights
(m, mean=0.0, std=0.01)
soulxpodcast/models/modules/hifigan_components/layers.py:20
Functionlifespan
应用生命周期管理
api/main.py:49
Functionload_wav
(full_path)
soulxpodcast/utils/audio.py:9
Methodposition_encoding
(self, offset: Union[int, torch.Tensor], size: int)
soulxpodcast/models/modules/flow_components/upsample_encoder.py:295
Functionremove_space_between_chinese
(text)
soulxpodcast/utils/text.py:4
Methodremove_weight_norm
(self)
soulxpodcast/models/modules/hifigan.py:173
Methodremove_weight_norm
(self)
soulxpodcast/models/modules/hifigan_components/layers.py:147
Functionroot
根路径
api/main.py:115
Methodrun_inference
在独立线程中执行推理
api/service.py:212
Methodset_chunk_feed_forward
(self, chunk_size: Optional[int], dim: int)
soulxpodcast/models/modules/flow_components/estimator.py:250
Functionsignal_handler
(signum, frame)
run_api.py:94
Functionspectral_de_normalize_torch
(magnitudes)
soulxpodcast/utils/audio.py:35
Functionupdate_example_choices
(dialect_key: str)
webui.py:344
Functionupdate_prompt_text
(dialect_key: str, example_key: str)
webui.py:355
Methodvalidate_dialogue_text
验证对话文本格式
api/models.py:61
← previous101–193 of 193, ranked by callers