update readme

README-en.md

@@ -38,6 +38,18 @@ You can then freely customize speakers under the `ComfyUI\models\TTS\Step-Audio-
 
 [2025-02-25]⚒️: Support custom speaker `custom_stpeaker`. 
 
+## Installation
+
+```
+cd ComfyUI/custom_nodes
+git clone https://github.com/billwuhao/ComfyUI_StepAudioTTS.git
+cd ComfyUI_StepAudioTTS
+pip install -r requirements.txt
+
+# python_embeded
+./python_embeded/python.exe -m pip install -r requirements.txt
+```
+
 ## Model Download
 
 Download to the `ComfyUI\models\TTS` folder

README.md

@@ -38,6 +38,17 @@ ComfyUI\models\TTS
 
 [2025-02-25]⚒️: 支持自定义说话者 `custom_speaker`. 
 
+## 安装
+
+```
+cd ComfyUI/custom_nodes
+git clone https://github.com/billwuhao/ComfyUI_StepAudioTTS.git
+cd ComfyUI_StepAudioTTS
+pip install -r requirements.txt
+
+# python_embeded
+./python_embeded/python.exe -m pip install -r requirements.txt
+```
 
 ## 模型下载
 

StepAudioTTS.py

@@ -7,6 +7,10 @@ import numpy as np
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from transformers.generation.logits_process import LogitsProcessor
 from transformers.generation.utils import LogitsProcessorList
+import sys
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, current_dir)
 
 from tokenizer import StepAudioTokenizer
 from cosyvoice.cli.cosyvoice import CosyVoice 

__init__.py

@@ -1,9 +1,3 @@
-import sys
-import os
-
-current_dir = os.path.dirname(os.path.abspath(__file__))
-sys.path.insert(0, current_dir)
-
-from StepAudioTTS import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
+from .StepAudioTTS import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
 
 __all__ = ["NODE_CLASS_MAPPINGS", "NODE_DISPLAY_NAME_MAPPINGS"]

funasr_detach/auto/auto_model.py

@@ -20,11 +20,6 @@ from funasr_detach.utils.load_utils import load_audio_text_image_video
 from funasr_detach.utils.timestamp_tools import timestamp_sentence
 from funasr_detach.models.campplus.utils import sv_chunk, postprocess, distribute_spk
 
-try:
-    from funasr_detach.models.campplus.cluster_backend import ClusterBackend
-except:
-    print("If you want to use the speaker diarization, please `pip install hdbscan`")
-
 
 def prepare_data_iterator(data_in, input_len=None, data_type=None, key=None):
     """
@@ -134,6 +129,12 @@ class AutoModel:
         # if spk_model is not None, build spk model else None
         spk_model = kwargs.get("spk_model", None)
         spk_kwargs = kwargs.get("spk_model_revision", None)
+                
+        try:
+            from funasr_detach.models.campplus.cluster_backend import ClusterBackend
+        except:
+            print("If you want to use the speaker diarization, please `pip install hdbscan`")
+
         if spk_model is not None:
             logging.info("Building SPK model.")
             spk_kwargs = {

funasr_detach/models/mossformer/mossformer_encoder.py

@@ -2,12 +2,6 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-try:
-    from rotary_embedding_torch import RotaryEmbedding
-except:
-    print(
-        "If you want use mossformer, please install rotary_embedding_torch by: \n pip install -U rotary_embedding_torch"
-    )
 from funasr_detach.models.transformer.layer_norm import (
     GlobalLayerNorm,
     CumulativeLayerNorm,
@@ -57,6 +51,12 @@ class MossformerBlock(nn.Module):
 
         self.group_size = group_size
 
+        try:
+            from rotary_embedding_torch import RotaryEmbedding
+        except:
+            print(
+                "If you want use mossformer, please install rotary_embedding_torch by: \n pip install -U rotary_embedding_torch"
+            )
         rotary_pos_emb = RotaryEmbedding(dim=min(32, query_key_dim))
         # max rotary embedding dimensions of 32, partial Rotary embeddings, from Wang et al - GPT-J
         self.layers = nn.ModuleList(

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "stepaudiotts_mw"
 description = "A Text To Speech node using Step-Audio-TTS in ComfyUI. Can speak, rap, sing, or clone voice."
-version = "1.1.4"
+version = "1.1.5"
 license = {file = "LICENSE"}
 dependencies = ["transformers>=4.48.3", "openai-whisper>=20231117", "sox>=1.5.0", "hyperpyyaml", "conformer>=0.3.2", "funasr>=1.1.3"]
 

requirements.txt

@@ -1,18 +1,14 @@
-# torch==2.3.1
-# torchaudio==2.3.1
-# torchvision==0.18.1
-# accelerate==1.3.0
-# onnxruntime-gpu==1.17.0
-# omegaconf==2.3.0
-# librosa==0.10.2.post1
-# modelscope
-# numpy==1.26.4
-# six==1.16.0
-# diffusers
-# pillow
-# sentencepiece
-# protobuf==5.29.3
-
+accelerate
+onnxruntime-gpu
+omegaconf
+librosa>=0.10.2.post1
+modelscope
+numpy
+six
+diffusers
+pillow
+sentencepiece
+protobuf
 transformers
 openai-whisper
 sox