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 callersFunctionresolve_attn_implementation
( requested: str | None, device: torch.device, dtype: torch.dtype, )
moss_tts_local_v1.5/streaming.py:105
↓ 1 callersFunctionresolve_attn_implementation
(requested: str, device: str, dtype: torch.dtype)
moss_tts_local_v1.5/finetuning/prepare_data.py:94
↓ 1 callersFunctionresolve_inference_asset
(model_path: str, filename: str)
moss_tts_realtime/finetuning/sft.py:284
↓ 1 callersFunctionresolve_inference_asset
(model_path: str, filename: str)
moss_tts_local/finetuning/sft.py:324
↓ 1 callersFunctionresolve_inference_asset
(model_path: str, filename: str)
moss_tts_delay/finetuning/sft.py:325
↓ 1 callersFunctionresolve_inference_asset
(model_path: str, filename: str)
moss_tts_local_v1.5/finetuning/sft.py:741
↓ 1 callersFunctionresolve_processor_token_strings
( fused_config: JSONDict, tokenizer_json_path: Path )
scripts/fuse_moss_tts_delay_with_codec.py:157
↓ 1 callersFunctionresolve_record_audio_paths
(record: Dict[str, Any], *, base_dir: Path)
moss_tts_local_v1.5/finetuning/common.py:129
↓ 1 callersFunctionresolve_shard_spec
( num_shards: Optional[int], shard_rank: Optional[int], default_num_shards: Optional[int] = None,
moss_tts_realtime/finetuning/common.py:76
↓ 1 callersFunctionresolve_shard_spec
( num_shards: Optional[int], shard_rank: Optional[int], default_num_shards: Optional[int] = None,
moss_tts_local/finetuning/common.py:76
↓ 1 callersFunctionresolve_shard_spec
( num_shards: Optional[int], shard_rank: Optional[int], default_num_shards: Optional[int] = None,
moss_tts_delay/finetuning/common.py:76
↓ 1 callersFunctionresolve_shard_spec
( num_shards: Optional[int], shard_rank: Optional[int], *, default_num_shards: Optional[int] =
moss_tts_local_v1.5/finetuning/common.py:69
↓ 1 callersFunctionresolve_torch_dtype
(mixed_precision: str)
moss_tts_realtime/finetuning/sft.py:141
↓ 1 callersFunctionresolve_torch_dtype
(mixed_precision: str)
moss_tts_local/finetuning/sft.py:153
↓ 1 callersFunctionresolve_torch_dtype
(mixed_precision: str)
moss_tts_delay/finetuning/sft.py:154
↓ 1 callersFunctionresolve_torch_dtype
(mixed_precision: str)
moss_tts_local_v1.5/finetuning/sft.py:248
↓ 1 callersFunctionresolve_warmup_steps
(args: argparse.Namespace, num_training_steps: int)
moss_tts_realtime/finetuning/sft.py:174
↓ 1 callersFunctionresolve_warmup_steps
(args: argparse.Namespace, num_training_steps: int)
moss_tts_local/finetuning/sft.py:186
↓ 1 callersFunctionresolve_warmup_steps
(args: argparse.Namespace, num_training_steps: int)
moss_tts_delay/finetuning/sft.py:187
↓ 1 callersFunctionresolve_warmup_steps
(args: argparse.Namespace, num_training_steps: int)
moss_tts_local_v1.5/finetuning/sft.py:298
↓ 1 callersMethodreturn_to_timestep
(self, timestep, sample, sample_stablized)
moss_soundeffect_v2/diffsynth/schedulers/flow_match.py:88
↓ 1 callersFunctionrotate_half
(hidden_states: torch.Tensor)
moss_tts_local_v1.5/gpt2_decoder.py:70
↓ 1 callersFunctionrun_batch
( pipeline: LlamaCppPipeline, cases: list[tuple[str, str, Path, str]], result_dir: Path, max_c
scripts/batch_eval_llama_cpp.py:109
↓ 1 callersFunctionrun_inference
( text: str, reference_audio: str | None, mode_with_reference: str, duration_control_enabled:
clis/moss_tts_app.py:339
↓ 1 callersFunctionrun_inference
( text: str, instruction: str, temperature: float, top_p: float, top_k: int, repetitio
clis/moss_voice_generator_app.py:146
↓ 1 callersFunctionrun_inference
(speaker_count: int, *all_inputs)
clis/moss_ttsd_app.py:396
↓ 1 callersFunctionrun_inference
( prompt: str, seconds: float, steps: int, cfg_scale: float, sigma_shift: float, seed:
clis/moss_sound_effect_app.py:37
↓ 1 callersFunctionrun_one_turn
( session: MossTTSRealtimeStreamingSession, codec, assistant_text: str, *, decode_chunk_fr
moss_tts_realtime/example_multiturn_stream_to_tts.py:166
↓ 1 callersMethodrun_stream
(self, request: StreamingRequest)
moss_tts_realtime/app.py:556
↓ 1 callersFunctionrun_streaming_tts
Receive stream text delta → push_text → output wav chunk in real time.
moss_tts_realtime/example_multiturn_stream_to_tts.py:115
↓ 1 callersFunctionrun_streaming_tts
Receives streaming text deltas, feeds them into the TTS via `session.push_text()`, and produces playable WAV chunks in real time. The pi
moss_tts_realtime/example_llm_stream_to_tts.py:139
↓ 1 callersFunctionrun_tests
(verbose: bool = True)
moss_tts_delay/tts_robust_normalizer_single_script.py:317
↓ 1 callersMethodsample
(self)
moss_soundeffect_v2/diffsynth/models/dac_vae.py:597
↓ 1 callersMethodsample_token
(self, logits, temperature, top_p=0.6, top_k=30, do_sample=True)
moss_tts_realtime/inferencer.py:411
↓ 1 callersMethodsample_token
(self, logits, temperature, top_p=0.6, top_k=30, do_sample=True)
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:431
↓ 1 callersFunctionsave_checkpoint
( accelerator: Accelerator, model: MossTTSRealtime, model_path: str, codec_path: str, outp
moss_tts_realtime/finetuning/sft.py:315
↓ 1 callersFunctionsave_checkpoint
( accelerator: Accelerator, model: MossTTSDelayModel, model_path: str, codec_path: str, ou
moss_tts_local/finetuning/sft.py:356
↓ 1 callersFunctionsave_checkpoint
( accelerator: Accelerator, model: MossTTSDelayModel, model_path: str, codec_path: str, ou
moss_tts_delay/finetuning/sft.py:357
↓ 1 callersMethodsave_training_state
Save the full training state to output_path/training_state/ for resume.
moss_soundeffect_v2/diffsynth/trainers/utils.py:816
↓ 1 callersFunctionselect_rank_shard
(records: Sequence[Dict[str, Any]], num_shards: int, shard_rank: int)
moss_tts_realtime/finetuning/common.py:91
↓ 1 callersFunctionselect_rank_shard
(records: Sequence[Dict[str, Any]], num_shards: int, shard_rank: int)
moss_tts_local/finetuning/common.py:91
↓ 1 callersFunctionselect_rank_shard
(records: Sequence[Dict[str, Any]], num_shards: int, shard_rank: int)
moss_tts_delay/finetuning/common.py:91
↓ 1 callersFunctionselect_rank_shard
(records: Sequence[Dict[str, Any]], num_shards: int, shard_rank: int)
moss_tts_local_v1.5/finetuning/common.py:91
↓ 1 callersFunctionset_gradient_checkpointing_flag
(module: torch.nn.Module, value: bool)
moss_tts_local_v1.5/finetuning/sft.py:527
↓ 1 callersFunctionshard_output_path
(path: str | Path, shard_rank: int, num_shards: int)
moss_tts_realtime/finetuning/common.py:95
↓ 1 callersFunctionshard_output_path
(path: str | Path, shard_rank: int, num_shards: int)
moss_tts_local/finetuning/common.py:95
↓ 1 callersFunctionshard_output_path
(path: str | Path, shard_rank: int, num_shards: int)
moss_tts_delay/finetuning/common.py:95
↓ 1 callersFunctionshard_output_path
(path: str | Path, shard_rank: int, num_shards: int)
moss_tts_local_v1.5/finetuning/common.py:95
↓ 1 callersFunctionshard_paths_for_rank
(paths: List[Path], world_size: int, rank: int)
moss_tts_realtime/finetuning/sft.py:344
↓ 1 callersFunctionshard_paths_for_rank
(paths: List[Path], world_size: int, rank: int)
moss_tts_local/finetuning/sft.py:385
↓ 1 callersFunctionshard_paths_for_rank
(paths: List[Path], world_size: int, rank: int)
moss_tts_delay/finetuning/sft.py:386
↓ 1 callersFunctionshard_paths_for_rank
(paths: List[Path], world_size: int, rank: int)
moss_tts_local_v1.5/finetuning/sft.py:807
↓ 1 callersFunctionsinusoidal_embedding_1d
(dim, position)
moss_soundeffect_v2/diffsynth/models/wan_video_dit.py:80
↓ 1 callersFunctionsnake
(x, alpha)
moss_soundeffect_v2/diffsynth/models/dac_vae.py:317
↓ 1 callersFunctionstart_session
( session_id: str, prompt_audio: str | None = None, assistant_text: str | None = None, user_te
moss_tts_realtime/tts_client.py:94
↓ 1 callersMethodstore
(self, hidden_states)
moss_soundeffect_v2/diffsynth/pipelines/wan_audio.py:621
↓ 1 callersMethodtext_encoder
(self)
moss_soundeffect_v2/pipeline_moss_soundeffect.py:64
↓ 1 callersMethodto
(self, *args, **kwargs)
moss_soundeffect_v2/pipeline_moss_soundeffect.py:87
↓ 1 callersMethodtrainable_modules
(self)
moss_soundeffect_v2/diffsynth/trainers/utils.py:734
↓ 1 callersMethodtrainable_param_names
(self)
moss_soundeffect_v2/diffsynth/trainers/utils.py:739
↓ 1 callersMethodtraining_loss
(self, **inputs)
moss_soundeffect_v2/diffsynth/pipelines/wan_audio.py:155
↓ 1 callersMethodtraining_target
(self, sample, noise, timestep)
moss_soundeffect_v2/diffsynth/schedulers/flow_match.py:114
↓ 1 callersMethodunpatchify
(self, x: torch.Tensor, grid_size: torch.Tensor)
moss_soundeffect_v2/diffsynth/models/wan_audio_dit.py:201
↓ 1 callersFunctionupdate_duration_controls
( enabled: bool, text: str, current_tokens: float | int | None, mode_with_reference: str, )
clis/moss_tts_app.py:197
↓ 1 callersFunctionupdate_speaker_panels
(speaker_count: int)
clis/moss_ttsd_app.py:254
↓ 1 callersMethodvae_output_to_image
(self, vae_output, pattern="B C H W", min_value=-1, max_value=1)
moss_soundeffect_v2/diffsynth/utils/__init__.py:75
↓ 1 callersMethodvalidate
(self)
moss_tts_delay/llama_cpp/pipeline.py:307
↓ 1 callersFunctionvalidate_args
(args: argparse.Namespace)
moss_tts_realtime/finetuning/sft.py:182
↓ 1 callersFunctionvalidate_args
(args: argparse.Namespace)
moss_tts_local/finetuning/sft.py:217
↓ 1 callersFunctionvalidate_args
(args: argparse.Namespace)
moss_tts_delay/finetuning/sft.py:218
↓ 1 callersFunctionvalidate_args
(args: argparse.Namespace)
moss_tts_local_v1.5/finetuning/sft.py:199
↓ 1 callersFunctionvalidate_args
(args: argparse.Namespace)
scripts/fuse_moss_tts_delay_with_codec.py:808
↓ 1 callersFunctionvalidate_records
Ensure every record has the required fields (conversations format).
moss_tts_realtime/finetuning/sft.py:209
↓ 1 callersFunctionverify_fused_artifact
( output_dir: Path, main_index: JSONDict, codec_index: JSONDict, prefix: str, )
scripts/fuse_moss_tts_delay_with_codec.py:363
↓ 1 callersFunctionverify_processor_artifacts
(output_dir: Path, fused_config: JSONDict)
scripts/fuse_moss_tts_delay_with_codec.py:668
↓ 1 callersMethodwait_for_capacity
(self, threshold_seconds: float)
moss_tts_realtime/app.py:248
↓ 1 callersFunctionwarmup_streaming_runtime
Pre-run small TTS and codec steps so the first user stream is hot.
moss_tts_local_v1.5/streaming.py:619
↓ 1 callersFunctionwrite_merged_config
(output_dir: Path, merged_config: JSONDict)
scripts/fuse_moss_tts_delay_with_codec.py:664
↓ 1 callersFunctionwrite_merged_index
(output_dir: Path, merged_weight_map: dict[str, str])
scripts/fuse_moss_tts_delay_with_codec.py:345
↓ 1 callersFunctionwrite_merged_shards
( output_dir: Path, merged_tensors: Iterable[tuple[str, torch.Tensor]], )
scripts/fuse_moss_tts_delay_with_codec.py:291
↓ 1 callersFunctionwrite_processor_config
(output_dir: Path, fused_config: JSONDict)
scripts/fuse_moss_tts_delay_with_codec.py:592
↓ 1 callersMethodwrite_sample
(self, data_id: int, sample: Dict[str, Any], compress: bool = True)
moss_soundeffect_v2/diffsynth/trainers/cache_shards.py:124
↓ 1 callersFunctionwrite_summary
(results: list[CaseResult], result_dir: Path)
scripts/batch_eval_llama_cpp.py:176
↓ 1 callersFunctionwrite_wav
(out_path: Path, sample_rate: int, chunks: Iterator[np.ndarray])
moss_tts_realtime/example_multiturn_stream_to_tts.py:148
↓ 1 callersFunctionwrite_wav
(out_path: Path, sample_rate: int, chunks: Iterator[np.ndarray])
moss_tts_realtime/example_llm_stream_to_tts.py:116
Method__call__
(self, *args, **kwargs)
moss_tts_local/processing_moss_tts.py:250
Method__call__
(self, *args, **kwargs)
moss_tts_delay/processing_moss_tts.py:249
Method__call__
( self, text: Union[ TextInput, PreTokenizedInput, list[TextInput], list[PreTokeni
moss_tts_delay/processing_moss_tts_delay_with_codec.py:74
Method__call__
Compute logits for all 33 heads. Returns: text_logits: (text_vocab_size,) or (B, text_vocab_size) audio_logits: (32,
moss_tts_delay/llama_cpp/lm_heads.py:50
Method__call__
(self, hidden_state: np.ndarray)
moss_tts_delay/llama_cpp/lm_heads.py:156
Method__call__
Compute summed embedding from multi-channel token IDs. Args: input_ids: (B, 1+N_VQ) or (B, S, 1+N_VQ) Returns:
moss_tts_delay/llama_cpp/embedding.py:50
Method__call__
(self, *args, **kwargs)
moss_tts_local_v1.5/processing_moss_tts.py:483
Method__call__
Run denoising and return a waveform of shape ``(B, C, T)``. Args: prompt: A single prompt or a batch of prompts. seco
moss_soundeffect_v2/pipeline_moss_soundeffect.py:156
Method__call__
(self, unit: PipelineUnit, pipe: BasePipeline, inputs_shared: dict, inputs_posi: dict, inputs_nega: dict)
moss_soundeffect_v2/diffsynth/utils/__init__.py:184
Method__call__
( self, # Prompt prompt: Union[str, list[str]], negative_prompt: Optional[Unio
moss_soundeffect_v2/diffsynth/pipelines/wan_audio.py:284
Method__call__
(self, sequence, **kwargs)
moss_soundeffect_v2/diffsynth/prompters/wan_prompter.py:47
Method__del__
(self)
moss_tts_delay/llama_cpp/backbone.py:235
Method__enter__
(self)
moss_tts_delay/llama_cpp/backbone.py:238
Method__enter__
(self)
moss_tts_local_v1.5/streaming.py:585
Method__enter__
(self)
scripts/fuse_moss_tts_delay_with_codec.py:54
← previousnext →801–900 of 1,298, ranked by callers