MCPcopy Create free account

hub / github.com/OpenMOSS/MOSS-TTS / functions

Functions1,298 in github.com/OpenMOSS/MOSS-TTS

⨍ Functions 1,298◇ Types & classes 178↳ Endpoints 16
↓ 1 callersFunctionload_source_index
Load safetensors weight map from model directory.
moss_tts_delay/llama_cpp/conversion/extract_weights.py:52
↓ 1 callersFunctionload_source_index
(model_dir: Path)
scripts/extract_weights_llama_cpp.py:45
↓ 1 callersFunctionload_state_dict
(file_path, torch_dtype=None, device="cpu")
moss_soundeffect_v2/diffsynth/models/utils.py:65
↓ 1 callersFunctionload_state_dict_from_bin
(file_path, torch_dtype=None, device="cpu")
moss_soundeffect_v2/diffsynth/models/utils.py:82
↓ 1 callersFunctionload_state_dict_from_safetensors
(file_path, torch_dtype=None, device="cpu")
moss_soundeffect_v2/diffsynth/models/utils.py:72
↓ 1 callersFunctionload_tokenizer_if_available
(accelerator: Accelerator)
community/norwegian-lora/train_lora.py:114
↓ 1 callersMethodload_training_metadata
(state_dir)
moss_soundeffect_v2/diffsynth/trainers/utils.py:838
↓ 1 callersFunctionload_training_model
( *, accelerator: Accelerator, model_path: str, model_dtype: torch.dtype, attn_implementat
moss_tts_local_v1.5/finetuning/sft.py:495
↓ 1 callersMethodload_video
(self, file_path)
moss_soundeffect_v2/diffsynth/trainers/utils.py:268
↓ 1 callersFunctionlora_grad_stats
(model: torch.nn.Module)
community/norwegian-lora/train_lora.py:400
↓ 1 callersMethodloudness_normalize
( wav: torch.Tensor, target_dbfs: float = -20, gain_range: tuple[float, float] = (-3.0
moss_tts_local/processing_moss_tts.py:754
↓ 1 callersMethodloudness_normalize
( wav: torch.Tensor, target_dbfs: float = -20, gain_range: tuple[float, float] = (-3.0
moss_tts_delay/processing_moss_tts.py:756
↓ 1 callersMethodloudness_normalize
( wav: np.ndarray, target_dbfs: float = -20, gain_range: tuple[float, float] = (-3.0,
moss_tts_delay/processing_moss_tts_delay_with_codec.py:236
↓ 1 callersMethodloudness_normalize
( wav: torch.Tensor, target_dbfs: float = -20.0, gain_range: tuple[float, float] = (-3
moss_tts_local_v1.5/processing_moss_tts.py:776
↓ 1 callersFunctionmain
()
clis/moss_tts_app.py:636
↓ 1 callersFunctionmain
()
clis/moss_voice_generator_app.py:367
↓ 1 callersFunctionmain
()
clis/moss_ttsd_app.py:765
↓ 1 callersFunctionmain
()
clis/moss_sound_effect_app.py:169
↓ 1 callersFunctionmain
()
clis/moss_tts_local_v1.5_app.py:1824
↓ 1 callersFunctionmain
(model_path, codec_path)
moss_tts_realtime/infer.py:15
↓ 1 callersFunctionmain
()
moss_tts_realtime/example_multiturn_stream_to_tts.py:251
↓ 1 callersFunctionmain
()
moss_tts_realtime/tts_client.py:148
↓ 1 callersFunctionmain
()
moss_tts_realtime/app.py:1343
↓ 1 callersFunctionmain
()
moss_tts_realtime/fast_api.py:992
↓ 1 callersFunctionmain
()
moss_tts_realtime/example_llm_stream_to_tts.py:187
↓ 1 callersFunctionmain
()
moss_tts_realtime/finetuning/sft.py:386
↓ 1 callersFunctionmain
()
moss_tts_realtime/finetuning/prepare_data.py:277
↓ 1 callersFunctionmain
()
moss_tts_local/finetuning/sft.py:427
↓ 1 callersFunctionmain
()
moss_tts_local/finetuning/prepare_data.py:152
↓ 1 callersFunctionmain
()
moss_tts_delay/finetuning/sft.py:428
↓ 1 callersFunctionmain
()
moss_tts_delay/finetuning/prepare_data.py:152
↓ 1 callersFunctionmain
()
moss_tts_delay/llama_cpp/conversion/extract_weights.py:263
↓ 1 callersFunctionmain
()
moss_tts_local_v1.5/streaming.py:1083
↓ 1 callersFunctionmain
()
moss_tts_local_v1.5/finetuning/sft.py:876
↓ 1 callersFunctionmain
()
moss_tts_local_v1.5/finetuning/prepare_data.py:245
↓ 1 callersFunctionmain
()
scripts/batch_eval_llama_cpp.py:226
↓ 1 callersFunctionmain
()
scripts/extract_weights_llama_cpp.py:241
↓ 1 callersFunctionmain
()
scripts/fuse_moss_tts_delay_with_codec.py:833
↓ 1 callersFunctionmain
()
moss_soundeffect_v2/infer_from_pipeline.py:18
↓ 1 callersFunctionmain
()
moss_soundeffect_v2/finetuning/export_to_hf.py:15
↓ 1 callersFunctionmain
()
moss_soundeffect_v2/finetuning/finetuning.py:235
↓ 1 callersFunctionmain
()
community/norwegian-lora/train_lora.py:416
↓ 1 callersMethodmake_user_prompt
(self, text: str, audio_tokens: np.ndarray)
moss_tts_realtime/mossttsrealtime/processing_mossttsrealtime.py:131
↓ 1 callersMethodmake_voice_clone_prompt
(self, prompt_audio_tokens_len)
moss_tts_realtime/inferencer.py:29
↓ 1 callersMethodmake_voice_clone_prompt
(self, prompt_audio_tokens_len: int)
moss_tts_realtime/mossttsrealtime/processing_mossttsrealtime.py:80
↓ 1 callersFunctionmodel_init_context
(accelerator: Accelerator)
moss_tts_realtime/finetuning/sft.py:260
↓ 1 callersFunctionmodel_init_context
(accelerator: Accelerator)
moss_tts_local/finetuning/sft.py:306
↓ 1 callersFunctionmodel_init_context
(accelerator: Accelerator)
moss_tts_delay/finetuning/sft.py:307
↓ 1 callersFunctionmodel_init_context
(accelerator: Accelerator)
moss_tts_local_v1.5/finetuning/sft.py:387
↓ 1 callersFunctionmoss_tts_masked_embedding
对 input_ids 中 != ignore_index 的位置做 embedding，ignore_index 的位置输出全 0 向量。 Args: embedding: 一个 nn.Embedding 层 input_ids: 任意形状的 L
moss_tts_local/modeling_moss_tts.py:97
↓ 1 callersFunctionmove_batch_to_device
(batch: Dict[str, Any], device: torch.device)
moss_tts_local_v1.5/finetuning/sft.py:851
↓ 1 callersFunctionnormalize_audio_code_list
( value: Any, field_name: str, allow_none: bool = False, )
moss_tts_local/finetuning/dataset.py:32
↓ 1 callersFunctionnormalize_audio_code_list
( value: Any, field_name: str, allow_none: bool = False, )
moss_tts_delay/finetuning/dataset.py:32
↓ 1 callersFunctionnormalize_language_for_tokens
(language: str, text: str)
moss_tts_local_v1.5/streaming.py:143
↓ 1 callersMethodon_epoch_end
(self, accelerator, model, epoch_id)
moss_soundeffect_v2/diffsynth/trainers/utils.py:790
↓ 1 callersMethodon_step_end
(self, accelerator, model, save_steps=None, loss=None)
moss_soundeffect_v2/diffsynth/trainers/utils.py:784
↓ 1 callersMethodon_training_end
(self, accelerator, model, save_steps=None)
moss_soundeffect_v2/diffsynth/trainers/utils.py:801
↓ 1 callersFunctionparse_args
()
moss_tts_realtime/finetuning/sft.py:78
↓ 1 callersFunctionparse_args
()
moss_tts_realtime/finetuning/prepare_data.py:40
↓ 1 callersFunctionparse_args
()
moss_tts_local/finetuning/sft.py:83
↓ 1 callersFunctionparse_args
()
moss_tts_local/finetuning/prepare_data.py:38
↓ 1 callersFunctionparse_args
()
moss_tts_delay/finetuning/sft.py:84
↓ 1 callersFunctionparse_args
()
moss_tts_delay/finetuning/prepare_data.py:38
↓ 1 callersFunctionparse_args
()
moss_tts_local_v1.5/finetuning/sft.py:94
↓ 1 callersFunctionparse_args
()
moss_tts_local_v1.5/finetuning/prepare_data.py:29
↓ 1 callersFunctionparse_args
()
scripts/fuse_moss_tts_delay_with_codec.py:699
↓ 1 callersFunctionparse_args
()
community/norwegian-lora/train_lora.py:37
↓ 1 callersFunctionparse_channelwise_loss_weight
(spec: Optional[str], n_heads: int)
moss_tts_local/finetuning/sft.py:194
↓ 1 callersFunctionparse_channelwise_loss_weight
(spec: Optional[str], n_heads: int)
moss_tts_delay/finetuning/sft.py:195
↓ 1 callersFunctionparse_channelwise_loss_weight
(spec: Optional[str], n_heads: int)
moss_tts_local_v1.5/finetuning/sft.py:306
↓ 1 callersFunctionparse_generation_output
Parse the generated output into text content and audio codes.
moss_tts_delay/llama_cpp/processor.py:223
↓ 1 callersFunctionpatch_tensor_constructor
(fn)
moss_soundeffect_v2/diffsynth/models/utils.py:26
↓ 1 callersFunctionpatch_tokenizer_config_processor_class
( tokenizer_config_path: Path, processor_class: str )
scripts/fuse_moss_tts_delay_with_codec.py:599
↓ 1 callersMethodpatchify
(self, x: torch.Tensor, control_camera_latents_input: Optional[torch.Tensor] = None)
moss_soundeffect_v2/diffsynth/models/wan_audio_dit.py:191
↓ 1 callersFunctionprecompute_freqs_cis_1d
(dim: int, end: int = 16384, theta: float = 10000.0)
moss_soundeffect_v2/diffsynth/models/wan_audio_dit.py:42
↓ 1 callersFunctionprecompute_freqs_cis_3d
(dim: int, end: int = 1024, theta: float = 10000.0)
moss_soundeffect_v2/diffsynth/models/wan_video_dit.py:87
↓ 1 callersMethodprefill
( self, input_ids, text_prefix_ids, max_prefill_len: Optional[int] = None,
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:153
↓ 1 callersMethodpreprocess_image
(self, image, torch_dtype=None, device=None, pattern="B C H W", min_value=-1, max_value=1)
moss_soundeffect_v2/diffsynth/utils/__init__.py:59
↓ 1 callersFunctionprocess_pose_file
(cam_params, width=672, height=384, original_pose_width=1280, original_pose_height=720, device='cpu', return_p
moss_soundeffect_v2/diffsynth/models/wan_video_camera_controller.py:150
↓ 1 callersMethodprocess_prompt
(self, prompt, positive=True)
moss_soundeffect_v2/diffsynth/prompters/base_prompter.py:57
↓ 1 callersFunctionprocessor_init_context
(accelerator: Accelerator)
moss_tts_realtime/finetuning/sft.py:240
↓ 1 callersFunctionprocessor_init_context
(accelerator: Accelerator)
moss_tts_local/finetuning/sft.py:286
↓ 1 callersFunctionprocessor_init_context
(accelerator: Accelerator)
moss_tts_delay/finetuning/sft.py:287
↓ 1 callersFunctionprocessor_init_context
(accelerator: Accelerator)
moss_tts_local_v1.5/finetuning/sft.py:343
↓ 1 callersFunctionprocessor_needs_audio_tokenizer
(records: List[Dict[str, Any]])
moss_tts_local/finetuning/sft.py:244
↓ 1 callersFunctionprocessor_needs_audio_tokenizer
(records: List[Dict[str, Any]])
moss_tts_delay/finetuning/sft.py:245
↓ 1 callersFunctionprocessor_needs_audio_tokenizer
(records: List[Dict[str, Any]])
moss_tts_local_v1.5/finetuning/sft.py:327
↓ 1 callersMethodpush_text_delta
Push a chunk of incremental text output from the LLM and return newly generated WAV chunks. Internally, this directly calls `session.
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:973
↓ 1 callersFunctionray_condition
Copied from https://github.com/hehao13/CameraCtrl/blob/main/inference.py
moss_soundeffect_v2/diffsynth/models/wan_video_camera_controller.py:114
↓ 1 callersMethodrelease
(self)
moss_soundeffect_v2/diffsynth/trainers/cache_shards.py:48
↓ 1 callersFunctionremap_backbone_name
Map a MossTTSDelay tensor name to Qwen3ForCausalLM convention. Returns None if the tensor does not belong to the backbone.
moss_tts_delay/llama_cpp/conversion/extract_weights.py:36
↓ 1 callersFunctionremap_backbone_name
Map a MossTTSDelay tensor name to Qwen3ForCausalLM convention.
scripts/extract_weights_llama_cpp.py:36
↓ 1 callersFunctionremap_main_tensor_name
(tensor_name: str)
scripts/fuse_moss_tts_delay_with_codec.py:241
↓ 1 callersFunctionrequire_json_array
(value: object, context: str)
scripts/fuse_moss_tts_delay_with_codec.py:120
↓ 1 callersFunctionresample_if_needed
(x: np.ndarray, src_sr: int, dst_sr: int)
moss_tts_realtime/fast_api.py:434
↓ 1 callersFunctionresolve_accelerate_mixed_precision
(mixed_precision: str)
moss_tts_local_v1.5/finetuning/sft.py:258
↓ 1 callersFunctionresolve_attn_implementation
()
moss_tts_realtime/infer.py:19
↓ 1 callersFunctionresolve_attn_implementation
(requested: str, dtype: torch.dtype)
moss_tts_realtime/finetuning/sft.py:151
↓ 1 callersFunctionresolve_attn_implementation
(requested: str, dtype: torch.dtype)
moss_tts_local/finetuning/sft.py:163
↓ 1 callersFunctionresolve_attn_implementation
(requested: str, dtype: torch.dtype)
moss_tts_delay/finetuning/sft.py:164
← previousnext →701–800 of 1,298, ranked by callers