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
↓ 2 callersFunctionwhitespace_clean
(text)
moss_soundeffect_v2/diffsynth/prompters/wan_prompter.py:16
↓ 1 callersFunctionWNConvTranspose1d
(*args, **kwargs)
moss_soundeffect_v2/diffsynth/models/dac_vae.py:311
↓ 1 callersMethod__enter__
(self)
moss_tts_delay/llama_cpp/pipeline.py:822
↓ 1 callersMethod__init__
( self, language_config: Qwen3Config | dict | None = None, local_config: MossTTSRealti
moss_tts_realtime/mossttsrealtime/configuration_mossttsrealtime.py:88
↓ 1 callersMethod__init__
(self, output_path, remove_prefix_in_ckpt=None, state_dict_converter=lambda x:x)
moss_soundeffect_v2/diffsynth/trainers/utils.py:777
↓ 1 callersMethod_apply_num_samples
(self, waveform)
moss_soundeffect_v2/diffsynth/trainers/utils.py:448
↓ 1 callersMethod_apply_repetition_penalty
( self, scores: torch.FloatTensor, previous_token_ids: Optional[torch.LongTensor],
moss_tts_local_v1.5/modeling_moss_tts.py:367
↓ 1 callersFunction_apply_repetition_penalty_with_seen_mask
( scores: torch.Tensor, seen_mask: Optional[torch.Tensor], penalty: float, )
moss_tts_local_v1.5/streaming.py:271
↓ 1 callersFunction_apply_seed
(seed: int | None)
moss_tts_realtime/app.py:50
↓ 1 callersFunction_attach_codes
Attach audio_codes to each conversation turn and ref_audio_codes to the record.
moss_tts_realtime/finetuning/prepare_data.py:261
↓ 1 callersMethod_audio_pad_token_id
(self)
moss_tts_local_v1.5/finetuning/dataset.py:76
↓ 1 callersMethod_autoregressive_loop
( self, input_ids: np.ndarray, max_new_tokens: int, backbone: LlamaCppBackbone
moss_tts_delay/llama_cpp/pipeline.py:681
↓ 1 callersFunction_base_cleanup
(text: str)
moss_tts_delay/tts_robust_normalizer_single_script.py:104
↓ 1 callersMethod_build_audio_tokenizer
(config: PipelineConfig)
moss_tts_delay/llama_cpp/pipeline.py:434
↓ 1 callersFunction_build_block
(length: int)
moss_tts_delay/llama_cpp/processor.py:155
↓ 1 callersMethod_build_continuation_codes
( self, conversation: list[Dict[str, Any]], n_vq: int, )
moss_tts_local_v1.5/processing_moss_tts.py:428
↓ 1 callersFunction_build_demo
( args: argparse.Namespace, tts_demo: StreamingTTSDemo, warmup_manager: WarmupManager, )
moss_tts_realtime/app.py:1146
↓ 1 callersMethod_build_generation_config
( self, generation_config: Optional[GenerationConfig] = None, max_new_tokens: Optional
moss_tts_local/modeling_moss_tts.py:642
↓ 1 callersMethod_build_generation_or_voice_clone_codes
( self, message: Dict[str, Any], n_vq: int, )
moss_tts_local_v1.5/processing_moss_tts.py:385
↓ 1 callersMethod_build_local_past_key_values
(self)
moss_tts_realtime/inferencer.py:93
↓ 1 callersMethod_build_local_past_key_values
(self)
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:96
↓ 1 callersMethod_build_prefill_batch
( self, input_ids_list: List[np.ndarray], text_ids_list: List[List[int]], text
moss_tts_realtime/inferencer.py:145
↓ 1 callersFunction_build_prefixed_text
( dialogue_text: str, prompt_text_map: dict[int, str], cloned_speakers: list[int], )
clis/moss_ttsd_app.py:348
↓ 1 callersFunction_build_preset_table_rows
()
clis/moss_ttsd_app.py:72
↓ 1 callersFunction_build_request
( args: argparse.Namespace, *, user_text: str | None, assistant_text: str | None, prompt_a
moss_tts_realtime/app.py:883
↓ 1 callersMethod_build_tied_weights_keys
(config: MossTTSLocalConfig)
moss_tts_local_v1.5/modeling_moss_tts.py:104
↓ 1 callersMethod_build_turn
Build input_ids [T_turn, 17] and labels [T_turn, 17] for one turn. Labels are set only when ``is_assistant`` is True. BOS (1025) is
moss_tts_realtime/finetuning/dataset.py:113
↓ 1 callersMethod_clean
(self, text)
moss_soundeffect_v2/diffsynth/prompters/wan_prompter.py:73
↓ 1 callersFunction_collect_all_wav_paths
Collect all unique wav paths from conversations and ref_wav fields.
moss_tts_realtime/finetuning/prepare_data.py:247
↓ 1 callersMethod_compute_input_embeddings
Computes the combined embeddings from text and multiple audio VQ channels. Args: input_ids: Shape (Batch, Seq_Len, 1 + n
moss_tts_delay/modeling_moss_tts.py:199
↓ 1 callersFunction_convert_hf_dit_state_dict
(state_dict: dict)
moss_soundeffect_v2/diffsynth/pipelines/wan_audio.py:100
↓ 1 callersFunction_copy_tree
(src: str, dst: str)
moss_soundeffect_v2/hf_export.py:78
↓ 1 callersFunction_detect_torch
()
moss_tts_delay/llama_cpp/pipeline.py:71
↓ 1 callersMethod_eager_attention
( self, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, a
moss_tts_local_v1.5/gpt2_decoder.py:166
↓ 1 callersMethod_eager_attention
( self, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, a
moss_tts_local_v1.5/qwen3_decoder.py:165
↓ 1 callersMethod_encode_audio_paths
(self, paths: List[str], target_n_vq: int)
moss_tts_local/finetuning/dataset.py:77
↓ 1 callersMethod_encode_audio_paths
(self, paths: List[str], target_n_vq: int)
moss_tts_delay/finetuning/dataset.py:77
↓ 1 callersMethod_encode_audio_paths
(self, paths: List[str], target_n_vq: int)
moss_tts_local_v1.5/finetuning/dataset.py:85
↓ 1 callersFunction_encode_batch_with_codec
Encode a batch of mono waveform tensors to a list of [T, NQ] int64 CPU tensors. Handles both the newer ``batch_encode`` API and the ``encode`` AP
moss_tts_realtime/finetuning/prepare_data.py:123
↓ 1 callersFunction_encode_chunk
(sr: int, chunk: np.ndarray, idx: int)
moss_tts_realtime/app.py:870
↓ 1 callersMethod_encode_reference_audio
(self, reference_audio_path: str, device: torch.device)
moss_tts_realtime/inferencer.py:114
↓ 1 callersFunction_encode_reference_audio_codes
( processor, clone_wavs: list[torch.Tensor], cloned_speakers: list[int], speaker_count: int,
clis/moss_ttsd_app.py:357
↓ 1 callersFunction_ensure_audio_resampled_file
(src: Path, target_sample_rate: int = SAMPLE_RATE)
moss_tts_realtime/fast_api.py:216
↓ 1 callersMethod_ensure_binary_idx
(self, idx_path: str, idx_bin_path: str)
moss_soundeffect_v2/diffsynth/trainers/cache_shards.py:200
↓ 1 callersMethod_ensure_max_num_samples
(self, waveform)
moss_soundeffect_v2/diffsynth/trainers/utils.py:462
↓ 1 callersMethod_ensure_min_num_samples
(self, waveform)
moss_soundeffect_v2/diffsynth/trainers/utils.py:474
↓ 1 callersMethod_ensure_mono
(self, waveform)
moss_soundeffect_v2/diffsynth/trainers/utils.py:431
↓ 1 callersMethod_ensure_warmup_text
(tokenizer, minimum_tokens: int)
moss_tts_realtime/app.py:711
↓ 1 callersMethod_extract_codec_codes
(self, encode_result: Any)
moss_tts_realtime/inferencer.py:106
↓ 1 callersFunction_extract_codes
(encode_result)
moss_tts_realtime/app.py:71
↓ 1 callersFunction_extract_codes
(encode_result)
moss_tts_realtime/fast_api.py:249
↓ 1 callersFunction_extract_codes
(encode_result)
moss_tts_realtime/example_llm_stream_to_tts.py:25
↓ 1 callersMethod_extract_text_segments
(self, force: bool)
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:652
↓ 1 callersMethod_filter_logits
( self, logits: torch.FloatTensor, top_k: Optional[int], top_p: Optional[float
moss_tts_local_v1.5/modeling_moss_tts.py:345
↓ 1 callersFunction_filter_records
Skip records that contain no assistant turn.
moss_tts_realtime/finetuning/prepare_data.py:229
↓ 1 callersFunction_find_bridge_lib
Locate the compiled bridge shared library.
moss_tts_delay/llama_cpp/backbone.py:22
↓ 1 callersFunction_find_last_equal
(arr: np.ndarray, value: int)
moss_tts_delay/llama_cpp/delay_state.py:307
↓ 1 callersFunction_find_last_equal
(input_ids: torch.LongTensor, value: int)
moss_tts_local_v1.5/modeling_moss_tts.py:29
↓ 1 callersFunction_find_latest_checkpoint
(output_path: str)
moss_soundeffect_v2/finetuning/finetuning.py:172
↓ 1 callersMethod_flash_attention
( self, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, a
moss_tts_local_v1.5/gpt2_decoder.py:225
↓ 1 callersMethod_flash_attention
( self, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, a
moss_tts_local_v1.5/qwen3_decoder.py:222
↓ 1 callersFunction_format_completion_status
( chunk_count: int, sample_rate: int, full_audio: np.ndarray, started_at: float, first_chu
moss_tts_realtime/app.py:386
↓ 1 callersMethod_generate_from_ids
( self, input_ids: List[np.ndarray], text_ids: List[List[int]], text_lengths
moss_tts_realtime/inferencer.py:196
↓ 1 callersMethod_get_local_transformer_runner
(self)
moss_tts_realtime/mossttsrealtime/streaming_mossttsrealtime.py:101
↓ 1 callersFunction_get_rank_and_world_size
()
moss_soundeffect_v2/diffsynth/trainers/cache_shards.py:18
↓ 1 callersMethod_get_ref_audio_codes
(self, record)
moss_tts_realtime/finetuning/dataset.py:98
↓ 1 callersFunction_get_unified_codes
Build the multi-channel (text + audio) packed sequence.
moss_tts_delay/llama_cpp/processor.py:167
↓ 1 callersMethod_get_unified_codes
此时的 content 已经是带上了对话格式
moss_tts_local/processing_moss_tts.py:563
↓ 1 callersMethod_get_unified_codes
此时的 content 已经是带上了对话格式
moss_tts_delay/processing_moss_tts.py:559
↓ 1 callersMethod_get_unified_codes
( self, role: str, content: str, audio_codes_list: list[torch.Tensor],
moss_tts_local_v1.5/processing_moss_tts.py:642
↓ 1 callersMethod_handle_finish_turn
(self, sess: SessionRuntime)
moss_tts_realtime/fast_api.py:813
↓ 1 callersMethod_handle_push_text
(self, sess: SessionRuntime, cmd: dict)
moss_tts_realtime/fast_api.py:795
↓ 1 callersMethod_handle_start_turn
(self, sess: SessionRuntime, cmd: dict)
moss_tts_realtime/fast_api.py:622
↓ 1 callersFunction_has_cjk
(text: str)
moss_tts_local_v1.5/streaming.py:139
↓ 1 callersFunction_html
(*, defaults: dict[str, Any], examples: list[dict[str, str]], languages: list[str], runtime: dict[str, Any])
clis/moss_tts_local_v1.5_app.py:701
↓ 1 callersMethod_init_or_validate_meta
(self)
moss_soundeffect_v2/diffsynth/trainers/cache_shards.py:82
↓ 1 callersFunction_load_and_resample
Load a single audio file as a mono waveform tensor resampled to `target_sr`.
moss_tts_realtime/finetuning/prepare_data.py:112
↓ 1 callersFunction_load_audio
(audio_path: str)
clis/moss_ttsd_app.py:164
↓ 1 callersFunction_load_audio
(path: Path, target_sample_rate: int = SAMPLE_RATE)
moss_tts_realtime/app.py:57
↓ 1 callersFunction_load_audio
(path: Path, target_sample_rate: int = SAMPLE_RATE)
moss_tts_realtime/fast_api.py:199
↓ 1 callersFunction_load_audio
(path: Path, target_sample_rate: int = 24000)
moss_tts_realtime/example_llm_stream_to_tts.py:29
↓ 1 callersMethod_load_audio
(self, audio_path: str, target_sample_rate: int)
moss_tts_realtime/inferencer.py:98
↓ 1 callersFunction_load_backend
( model_path: str, tokenizer_path: str, codec_model_path: str, device_str: str, attn_impl:
moss_tts_realtime/app.py:409
↓ 1 callersFunction_load_backend
( model_path: str, tokenizer_path: str, codec_model_path: str, device_str: str, attn_impl:
moss_tts_realtime/fast_api.py:272
↓ 1 callersFunction_load_bridge
Load the C bridge and set up function signatures.
moss_tts_delay/llama_cpp/backbone.py:38
↓ 1 callersFunction_load_codec
(codec_path: str, device: torch.device)
moss_tts_realtime/example_multiturn_stream_to_tts.py:38
↓ 1 callersFunction_load_codec
(device: torch.device, codec_model_path: str)
moss_tts_realtime/app.py:66
↓ 1 callersFunction_load_codec
(device: torch.device, codec_model_path: str)
moss_tts_realtime/fast_api.py:244
↓ 1 callersFunction_load_codec
(codec_path: str, device: torch.device)
moss_tts_realtime/example_llm_stream_to_tts.py:38
↓ 1 callersMethod_load_decoder_only
(self)
moss_tts_delay/llama_cpp/pipeline.py:465
↓ 1 callersFunction_load_defaults
()
moss_tts_delay/llama_cpp/_constants.py:36
↓ 1 callersMethod_load_encoder_only
(self)
moss_tts_delay/llama_cpp/pipeline.py:459
↓ 1 callersMethod_load_llm_components
Load backbone + embeddings + LM heads (low-memory mode).
moss_tts_delay/llama_cpp/pipeline.py:596
↓ 1 callersFunction_load_onnx_decoder
(onnx_path: str, use_gpu: bool)
moss_tts_delay/llama_cpp/pipeline.py:180
↓ 1 callersFunction_load_onnx_encoder
(onnx_path: str, use_gpu: bool)
moss_tts_delay/llama_cpp/pipeline.py:149
↓ 1 callersMethod_load_reference_wav
Load and normalize reference audio to a float32 waveform.
moss_tts_delay/llama_cpp/pipeline.py:646
↓ 1 callersFunction_load_trt_decoder
(engine_path: str)
moss_tts_delay/llama_cpp/pipeline.py:126
↓ 1 callersFunction_load_trt_encoder
(engine_path: str)
moss_tts_delay/llama_cpp/pipeline.py:95
↓ 1 callersMethod_local_past_length
(past_key_values: Optional[tuple[Any, ...]])
moss_tts_local_v1.5/modeling_moss_tts.py:197
↓ 1 callersFunction_looks_like_uri
(path: str)
moss_tts_local_v1.5/finetuning/common.py:169
← previousnext →301–400 of 1,298, ranked by callers