Merge branch 'main' into batching_infer

README.md

@@ -33,6 +33,8 @@ Unseen speakers few-shot fine-tuning demo:
 
 https://github.com/RVC-Boss/GPT-SoVITS/assets/129054828/05bee1fa-bdd8-4d85-9350-80c060ab47fb
 
+[教程中文版](https://www.yuque.com/baicaigongchang1145haoyuangong/ib3g1e) [User guide (EN)](https://rentry.co/GPT-SoVITS-guide#/)
+
 ## Installation
 
 For users in China region, you can [click here](https://www.codewithgpu.com/i/RVC-Boss/GPT-SoVITS/GPT-SoVITS-Official) to use AutoDL Cloud Docker to experience the full functionality online.
@@ -173,7 +175,7 @@ D:\GPT-SoVITS\xxx/xxx.wav|xxx|en|I like playing Genshin.
 - [ ] **High Priority:**
 
   - [x] Localization in Japanese and English.
-  - [ ] User guide.
+  - [x] User guide.
   - [x] Japanese and English dataset fine tune training.
 
 - [ ] **Features:**
@@ -218,24 +220,33 @@ ASR processing is performed through Faster_Whisper(ASR marking except Chinese)
 python ./tools/damo_asr/WhisperASR.py -i <input> -o <output> -f <file_name.list> -l <language>
 ```
 A custom list save path is enabled
+
 ## Credits
 
 Special thanks to the following projects and contributors:
 
+### Theoretical
 - [ar-vits](https://github.com/innnky/ar-vits)
 - [SoundStorm](https://github.com/yangdongchao/SoundStorm/tree/master/soundstorm/s1/AR)
 - [vits](https://github.com/jaywalnut310/vits)
 - [TransferTTS](https://github.com/hcy71o/TransferTTS/blob/master/models.py#L556)
-- [Chinese Speech Pretrain](https://github.com/TencentGameMate/chinese_speech_pretrain)
 - [contentvec](https://github.com/auspicious3000/contentvec/)
 - [hifi-gan](https://github.com/jik876/hifi-gan)
-- [Chinese-Roberta-WWM-Ext-Large](https://huggingface.co/hfl/chinese-roberta-wwm-ext-large)
 - [fish-speech](https://github.com/fishaudio/fish-speech/blob/main/tools/llama/generate.py#L41)
+### Pretrained Models
+- [Chinese Speech Pretrain](https://github.com/TencentGameMate/chinese_speech_pretrain)
+- [Chinese-Roberta-WWM-Ext-Large](https://huggingface.co/hfl/chinese-roberta-wwm-ext-large)
+### Text Frontend for Inference
+- [paddlespeech zh_normalization](https://github.com/PaddlePaddle/PaddleSpeech/tree/develop/paddlespeech/t2s/frontend/zh_normalization)
+- [LangSegment](https://github.com/juntaosun/LangSegment)
+### WebUI Tools
 - [ultimatevocalremovergui](https://github.com/Anjok07/ultimatevocalremovergui)
 - [audio-slicer](https://github.com/openvpi/audio-slicer)
 - [SubFix](https://github.com/cronrpc/SubFix)
 - [FFmpeg](https://github.com/FFmpeg/FFmpeg)
 - [gradio](https://github.com/gradio-app/gradio)
+- [faster-whisper](https://github.com/SYSTRAN/faster-whisper)
+- [FunASR](https://github.com/alibaba-damo-academy/FunASR)
   
 ## Thanks to all contributors for their efforts
 

docs/cn/Changelog_CN.md

@@ -127,7 +127,7 @@
 
 ### 20240221更新
 
-1-数据处理添加语音降噪选项
+1-数据处理添加语音降噪选项（降噪为只剩16k采样率，除非底噪很大先不急着用哦。）
 
 2-中文日文前端处理优化 https://github.com/RVC-Boss/GPT-SoVITS/pull/559 https://github.com/RVC-Boss/GPT-SoVITS/pull/556 https://github.com/RVC-Boss/GPT-SoVITS/pull/532 https://github.com/RVC-Boss/GPT-SoVITS/pull/507 https://github.com/RVC-Boss/GPT-SoVITS/pull/509
 
@@ -135,9 +135,22 @@
 
 4-colab修复不开启公网url
 
+### 20240306更新
+
+1-推理加速50%（RTX3090+pytorch2.2.1+cu11.8+win10+py39 tested）https://github.com/RVC-Boss/GPT-SoVITS/pull/672
+
+2-如果用faster whisper非中文ASR不再需要先下中文funasr模型
+
+3-修复uvr5去混响模型 是否混响 反的 https://github.com/RVC-Boss/GPT-SoVITS/pull/610
+
+4-faster whisper如果无cuda可用自动cpu推理 https://github.com/RVC-Boss/GPT-SoVITS/pull/675
+
+5-修改is_half的判断使在Mac上能正常CPU推理 https://github.com/RVC-Boss/GPT-SoVITS/pull/573
+
+
 todolist：
 
-1-中文多音字推理优化
+1-中文多音字推理优化(有没有人来测试的，欢迎把测试结果写在pr评论区里) https://github.com/RVC-Boss/GPT-SoVITS/pull/488
 
 
 

tools/asr/fasterwhisper_asr.py

@@ -4,12 +4,12 @@ os.environ["HF_ENDPOINT"]="https://hf-mirror.com"
 import traceback
 import requests
 from glob import glob
+import torch
 
 from faster_whisper import WhisperModel
 from tqdm import tqdm
 
 from tools.asr.config import check_fw_local_models
-from tools.asr.funasr_asr import only_asr
 
 os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
 
@@ -45,8 +45,9 @@ def execute_asr(input_folder, output_folder, model_size, language,precision):
     if language == 'auto':
         language = None #不设置语种由模型自动输出概率最高的语种
     print("loading faster whisper model:",model_size,model_path)
+    device = 'cuda' if torch.cuda.is_available() else 'cpu'
     try:
-        model = WhisperModel(model_path, device="cuda", compute_type=precision)
+        model = WhisperModel(model_path, device=device, compute_type=precision)
     except:
         return print(traceback.format_exc())
     output = []
@@ -68,6 +69,8 @@ def execute_asr(input_folder, output_folder, model_size, language,precision):
 
             if info.language == "zh":
                 print("检测为中文文本,转funasr处理")
+                if("only_asr"not in globals()):
+                    from tools.asr.funasr_asr import only_asr##如果用英文就不需要导入下载模型
                 text = only_asr(file)
 
             if text == '':