import os,sys from .ft_node import SoVITS_weight_root,GPT_weight_root,\ pretrained_sovits_name,pretrained_gpt_name,work_path,\ now_dir,gsv_path,models_dir,output_dir sys.path.append(now_dir) sys.path.append(gsv_path) from tools.i18n.i18n import I18nAuto, scan_language_list language=os.environ.get("language","Auto") language=sys.argv[-1] if sys.argv[-1] in scan_language_list() else language i18n = I18nAuto(language=language) def get_weights_names(GPT_weight_root, SoVITS_weight_root): SoVITS_names = [i for i in pretrained_sovits_name] for path in SoVITS_weight_root: try: for name in os.listdir(os.path.join(work_path,path)): if name.endswith(".pth"): SoVITS_names.append("%s/%s" % (path, name)) except: pass GPT_names = [i for i in pretrained_gpt_name] for path in GPT_weight_root: try: for name in os.listdir(os.path.join(work_path,path)): if name.endswith(".ckpt"): GPT_names.append("%s/%s" % (path, name)) except: pass return SoVITS_names, GPT_names dict_language_v1 = { i18n("中文"): "all_zh",#全部按中文识别 i18n("英文"): "en",#全部按英文识别#######不变 i18n("日文"): "all_ja",#全部按日文识别 i18n("中英混合"): "zh",#按中英混合识别####不变 i18n("日英混合"): "ja",#按日英混合识别####不变 i18n("多语种混合"): "auto",#多语种启动切分识别语种 } dict_language_v2 = { i18n("中文"): "all_zh",#全部按中文识别 i18n("英文"): "en",#全部按英文识别#######不变 i18n("日文"): "all_ja",#全部按日文识别 i18n("粤语"): "all_yue",#全部按中文识别 i18n("韩文"): "all_ko",#全部按韩文识别 i18n("中英混合"): "zh",#按中英混合识别####不变 i18n("日英混合"): "ja",#按日英混合识别####不变 i18n("粤英混合"): "yue",#按粤英混合识别####不变 i18n("韩英混合"): "ko",#按韩英混合识别####不变 i18n("多语种混合"): "auto",#多语种启动切分识别语种 i18n("多语种混合(粤语)"): "auto_yue",#多语种启动切分识别语种 } dict_language = dict_language_v2 import re import torch import numpy as np from time import time as ttime import cuda_malloc device = "cuda" if cuda_malloc.cuda_malloc_supported() else "cpu" from module.models import SynthesizerTrn splits = {"，", "。", "？", "！", ",", ".", "?", "!", "~", ":", "：", "—", "…", } def get_first(text): pattern = "[" + "".join(re.escape(sep) for sep in splits) + "]" text = re.split(pattern, text)[0].strip() return text class DictToAttrRecursive(dict): def __init__(self, input_dict): super().__init__(input_dict) for key, value in input_dict.items(): if isinstance(value, dict): value = DictToAttrRecursive(value) self[key] = value setattr(self, key, value) def __getattr__(self, item): try: return self[item] except KeyError: raise AttributeError(f"Attribute {item} not found") def __setattr__(self, key, value): if isinstance(value, dict): value = DictToAttrRecursive(value) super(DictToAttrRecursive, self).__setitem__(key, value) super().__setattr__(key, value) def __delattr__(self, item): try: del self[item] except KeyError: raise AttributeError(f"Attribute {item} not found") import GPT_SoVITS.utils as utils def change_sovits_weights(sovits_path): global vq_model, hps, version, dict_language comfyui_utils = sys.modules['utils'] sys.modules['utils'] = utils dict_s2 = torch.load(sovits_path, map_location="cpu") sys.modules['utils'] = comfyui_utils hps = dict_s2["config"] hps = DictToAttrRecursive(hps) hps.model.semantic_frame_rate = "25hz" if dict_s2['weight']['enc_p.text_embedding.weight'].shape[0] == 322: hps.model.version = "v1" else: hps.model.version = "v2" version = hps.model.version # print("sovits版本:",hps.model.version) vq_model = SynthesizerTrn( hps.data.filter_length // 2 + 1, hps.train.segment_size // hps.data.hop_length, n_speakers=hps.data.n_speakers, **hps.model ) if ("pretrained" not in sovits_path): del vq_model.enc_q if is_half == True: vq_model = vq_model.half().to(device) else: vq_model = vq_model.to(device) vq_model.eval() print(vq_model.load_state_dict(dict_s2["weight"], strict=False)) dict_language = dict_language_v1 if version =='v1' else dict_language_v2 ''' with open("./weight.json")as f: data=f.read() data=json.loads(data) data["SoVITS"][version]=sovits_path with open("./weight.json","w")as f:f.write(json.dumps(data)) ''' from AR.models.t2s_lightning_module import Text2SemanticLightningModule def change_gpt_weights(gpt_path): global hz, max_sec, t2s_model, config hz = 50 dict_s1 = torch.load(gpt_path, map_location="cpu") config = dict_s1["config"] max_sec = config["data"]["max_sec"] t2s_model = Text2SemanticLightningModule(config, "****", is_train=False) t2s_model.load_state_dict(dict_s1["weight"]) if is_half == True: t2s_model = t2s_model.half() t2s_model = t2s_model.to(device) t2s_model.eval() total = sum([param.nelement() for param in t2s_model.parameters()]) print("Number of parameter: %.2fM" % (total / 1e6)) ''' with open("./weight.json")as f: data=f.read() data=json.loads(data) data["GPT"][version]=gpt_path with open("./weight.json","w")as f:f.write(json.dumps(data)) ''' def process_text(texts): _text=[] if all(text in [None, " ", "\n",""] for text in texts): raise ValueError(i18n("请输入有效文本")) for text in texts: if text in [None, " ", ""]: pass else: _text.append(text) return _text from text import chinese,cleaned_text_to_sequence from text.cleaner import clean_text import LangSegment from module.mel_processing import spectrogram_torch from tools.my_utils import load_audio def get_spepc(hps, audio): # audio = load_audio(filename, int(hps.data.sampling_rate)) # audio = torch.FloatTensor(audio) maxx=audio.abs().max() if(maxx>1):audio/=min(2,maxx) audio_norm = audio audio_norm = audio_norm.unsqueeze(0) spec = spectrogram_torch( audio_norm, hps.data.filter_length, hps.data.sampling_rate, hps.data.hop_length, hps.data.win_length, center=False, ) return spec def clean_text_inf(text, language, version): phones, word2ph, norm_text = clean_text(text, language, version) phones = cleaned_text_to_sequence(phones, version) return phones, word2ph, norm_text def get_bert_feature(text, word2ph): with torch.no_grad(): inputs = tokenizer(text, return_tensors="pt") for i in inputs: inputs[i] = inputs[i].to(device) res = bert_model(**inputs, output_hidden_states=True) res = torch.cat(res["hidden_states"][-3:-2], -1)[0].cpu()[1:-1] assert len(word2ph) == len(text) phone_level_feature = [] for i in range(len(word2ph)): repeat_feature = res[i].repeat(word2ph[i], 1) phone_level_feature.append(repeat_feature) phone_level_feature = torch.cat(phone_level_feature, dim=0) return phone_level_feature.T def get_bert_inf(phones, word2ph, norm_text, language): language=language.replace("all_","") if language == "zh": bert = get_bert_feature(norm_text, word2ph).to(device)#.to(dtype) else: bert = torch.zeros( (1024, len(phones)), dtype=torch.float16 if is_half == True else torch.float32, ).to(device) return bert def get_phones_and_bert(text,language,version): if language in {"en", "all_zh", "all_ja", "all_ko", "all_yue"}: language = language.replace("all_","") if language == "en": LangSegment.setfilters(["en"]) formattext = " ".join(tmp["text"] for tmp in LangSegment.getTexts(text)) else: # 因无法区别中日韩文汉字,以用户输入为准 formattext = text while " " in formattext: formattext = formattext.replace(" ", " ") if language == "zh": if re.search(r'[A-Za-z]', formattext): formattext = re.sub(r'[a-z]', lambda x: x.group(0).upper(), formattext) formattext = chinese.mix_text_normalize(formattext) return get_phones_and_bert(formattext,"zh",version) else: phones, word2ph, norm_text = clean_text_inf(formattext, language, version) bert = get_bert_feature(norm_text, word2ph).to(device) elif language == "yue" and re.search(r'[A-Za-z]', formattext): formattext = re.sub(r'[a-z]', lambda x: x.group(0).upper(), formattext) formattext = chinese.mix_text_normalize(formattext) return get_phones_and_bert(formattext,"yue",version) else: phones, word2ph, norm_text = clean_text_inf(formattext, language, version) bert = torch.zeros( (1024, len(phones)), dtype=torch.float16 if is_half == True else torch.float32, ).to(device) elif language in {"zh", "ja", "ko", "yue", "auto", "auto_yue"}: textlist=[] langlist=[] LangSegment.setfilters(["zh","ja","en","ko"]) if language == "auto": for tmp in LangSegment.getTexts(text): langlist.append(tmp["lang"]) textlist.append(tmp["text"]) elif language == "auto_yue": for tmp in LangSegment.getTexts(text): if tmp["lang"] == "zh": tmp["lang"] = "yue" langlist.append(tmp["lang"]) textlist.append(tmp["text"]) else: for tmp in LangSegment.getTexts(text): if tmp["lang"] == "en": langlist.append(tmp["lang"]) else: # 因无法区别中日韩文汉字,以用户输入为准 langlist.append(language) textlist.append(tmp["text"]) print(textlist) print(langlist) phones_list = [] bert_list = [] norm_text_list = [] for i in range(len(textlist)): lang = langlist[i] phones, word2ph, norm_text = clean_text_inf(textlist[i], lang, version) bert = get_bert_inf(phones, word2ph, norm_text, lang) phones_list.append(phones) norm_text_list.append(norm_text) bert_list.append(bert) bert = torch.cat(bert_list, dim=1) phones = sum(phones_list, []) norm_text = ''.join(norm_text_list) dtype=torch.float16 if is_half == True else torch.float32 return phones,bert.to(dtype),norm_text def merge_short_text_in_array(texts, threshold): if (len(texts)) < 2: return texts result = [] text = "" for ele in texts: text += ele if len(text) >= threshold: result.append(text) text = "" if (len(text) > 0): if len(result) == 0: result.append(text) else: result[len(result) - 1] += text return result def get_tts_wav(ref_wav,prompt_text, prompt_language, text, text_language, how_to_cut=i18n("不切"), top_k=20, top_p=0.6, temperature=0.6, speed=1): t= [] if len(prompt_text) == 0: ref_free = True else: ref_free = False t0 = ttime() prompt_language = dict_language[prompt_language] text_language = dict_language[text_language] print(f"prompt_language:{prompt_language}") print(f"text_language:{text_language}") if not ref_free: prompt_text = prompt_text.strip("\n") if (prompt_text[-1] not in splits): prompt_text += "。" if prompt_language != "en" else "." print(i18n("实际输入的参考文本:"), prompt_text) text = text.strip("\n") if (text[0] not in splits and len(get_first(text)) < 4): text = "。" + text if text_language != "en" else "." + text print(i18n("实际输入的目标文本:"), text) zero_wav = np.zeros( int(hps.data.sampling_rate * 0.3), dtype=np.float16 if is_half == True else np.float32, ) if not ref_free: with torch.no_grad(): ''' wav16k, sr = librosa.load(ref_wav_path, sr=16000) if (wav16k.shape[0] > 160000 or wav16k.shape[0] < 48000): gr.Warning(i18n("参考音频在3~10秒范围外，请更换！")) raise OSError(i18n("参考音频在3~10秒范围外，请更换！")) wav16k = torch.from_numpy(wav16k) ''' wav16k = torchaudio.transforms.Resample(orig_freq=32000, new_freq=16000)(ref_wav) zero_wav_torch = torch.from_numpy(zero_wav) if is_half == True: wav16k = wav16k.half().to(device) zero_wav_torch = zero_wav_torch.half().to(device) else: wav16k = wav16k.to(device) zero_wav_torch = zero_wav_torch.to(device) wav16k = torch.cat([wav16k, zero_wav_torch]) ssl_content = ssl_model.model(wav16k.unsqueeze(0))[ "last_hidden_state" ].transpose( 1, 2 ) # .float() codes = vq_model.extract_latent(ssl_content) prompt_semantic = codes[0, 0] prompt = prompt_semantic.unsqueeze(0).to(device) t1 = ttime() t.append(t1-t0) if (how_to_cut == i18n("凑四句一切")): text = cut1(text) elif (how_to_cut == i18n("凑50字一切")): text = cut2(text) elif (how_to_cut == i18n("按中文句号。切")): text = cut3(text) elif (how_to_cut == i18n("按英文句号.切")): text = cut4(text) elif (how_to_cut == i18n("按标点符号切")): text = cut5(text) while "\n\n" in text: text = text.replace("\n\n", "\n") print(i18n("实际输入的目标文本(切句后):"), text) texts = text.split("\n") texts = process_text(texts) texts = merge_short_text_in_array(texts, 5) audio_opt = [] if not ref_free: phones1,bert1,norm_text1=get_phones_and_bert(prompt_text, prompt_language, version) for i_text,text in enumerate(texts): # 解决输入目标文本的空行导致报错的问题 if (len(text.strip()) == 0): continue if (text[-1] not in splits): text += "。" if text_language != "en" else "." print(i18n("实际输入的目标文本(每句):"), text) phones2,bert2,norm_text2=get_phones_and_bert(text, text_language, version) print(i18n("前端处理后的文本(每句):"), norm_text2) if not ref_free: bert = torch.cat([bert1, bert2], 1) all_phoneme_ids = torch.LongTensor(phones1+phones2).to(device).unsqueeze(0) else: bert = bert2 all_phoneme_ids = torch.LongTensor(phones2).to(device).unsqueeze(0) bert = bert.to(device).unsqueeze(0) all_phoneme_len = torch.tensor([all_phoneme_ids.shape[-1]]).to(device) t2 = ttime() # cache_key="%s-%s-%s-%s-%s-%s-%s-%s"%(ref_wav_path,prompt_text,prompt_language,text,text_language,top_k,top_p,temperature) # print(cache.keys(),if_freeze) ''' if(i_text in cache and if_freeze==True):pred_semantic=cache[i_text] else: with torch.no_grad(): pred_semantic, idx = t2s_model.model.infer_panel( all_phoneme_ids, all_phoneme_len, None if ref_free else prompt, bert, # prompt_phone_len=ph_offset, top_k=top_k, top_p=top_p, temperature=temperature, early_stop_num=hz * max_sec, ) pred_semantic = pred_semantic[:, -idx:].unsqueeze(0) cache[i_text]=pred_semantic ''' with torch.no_grad(): pred_semantic, idx = t2s_model.model.infer_panel( all_phoneme_ids, all_phoneme_len, None if ref_free else prompt, bert, # prompt_phone_len=ph_offset, top_k=top_k, top_p=top_p, temperature=temperature, early_stop_num=hz * max_sec, ) pred_semantic = pred_semantic[:, -idx:].unsqueeze(0) t3 = ttime() refers=[] ''' if(inp_refs): for path in inp_refs: try: refer = get_spepc(hps, path.name).to(dtype).to(device) refers.append(refer) except: traceback.print_exc() ''' dtype=torch.float16 if is_half == True else torch.float32 if(len(refers)==0):refers = [get_spepc(hps, ref_wav).to(dtype).to(device)] audio = (vq_model.decode(pred_semantic, torch.LongTensor(phones2).to(device).unsqueeze(0), refers,speed=speed).detach().cpu().numpy()[0, 0]) max_audio=np.abs(audio).max()#简单防止16bit爆音 if max_audio>1:audio/=max_audio audio_opt.append(audio) audio_opt.append(zero_wav) t4 = ttime() t.extend([t2 - t1,t3 - t2, t4 - t3]) t1 = ttime() print("%.3f\t%.3f\t%.3f\t%.3f" % (t[0], sum(t[1::3]), sum(t[2::3]), sum(t[3::3])) ) return torch.Tensor(np.concatenate(audio_opt, 0)).unsqueeze(0) ''' yield hps.data.sampling_rate, (np.concatenate(audio_opt, 0) * 32768).astype( np.int16 ) ''' punctuation = set(['!', '?', '…', ',', '.', '-'," "]) def split(todo_text): todo_text = todo_text.replace("……", "。").replace("——", "，") if todo_text[-1] not in splits: todo_text += "。" i_split_head = i_split_tail = 0 len_text = len(todo_text) todo_texts = [] while 1: if i_split_head >= len_text: break # 结尾一定有标点，所以直接跳出即可，最后一段在上次已加入 if todo_text[i_split_head] in splits: i_split_head += 1 todo_texts.append(todo_text[i_split_tail:i_split_head]) i_split_tail = i_split_head else: i_split_head += 1 return todo_texts def cut1(inp): inp = inp.strip("\n") inps = split(inp) split_idx = list(range(0, len(inps), 4)) split_idx[-1] = None if len(split_idx) > 1: opts = [] for idx in range(len(split_idx) - 1): opts.append("".join(inps[split_idx[idx]: split_idx[idx + 1]])) else: opts = [inp] opts = [item for item in opts if not set(item).issubset(punctuation)] return "\n".join(opts) def cut2(inp): inp = inp.strip("\n") inps = split(inp) if len(inps) < 2: return inp opts = [] summ = 0 tmp_str = "" for i in range(len(inps)): summ += len(inps[i]) tmp_str += inps[i] if summ > 50: summ = 0 opts.append(tmp_str) tmp_str = "" if tmp_str != "": opts.append(tmp_str) # print(opts) if len(opts) > 1 and len(opts[-1]) < 50: ##如果最后一个太短了，和前一个合一起 opts[-2] = opts[-2] + opts[-1] opts = opts[:-1] opts = [item for item in opts if not set(item).issubset(punctuation)] return "\n".join(opts) def cut3(inp): inp = inp.strip("\n") opts = ["%s" % item for item in inp.strip("。").split("。")] opts = [item for item in opts if not set(item).issubset(punctuation)] return "\n".join(opts) def cut4(inp): inp = inp.strip("\n") opts = ["%s" % item for item in inp.strip(".").split(".")] opts = [item for item in opts if not set(item).issubset(punctuation)] return "\n".join(opts) # contributed by https://github.com/AI-Hobbyist/GPT-SoVITS/blob/main/GPT_SoVITS/inference_webui.py def cut5(inp): inp = inp.strip("\n") punds = {',', '.', ';', '?', '!', '、', '，', '。', '？', '！', ';', '：', '…'} mergeitems = [] items = [] for i, char in enumerate(inp): if char in punds: if char == '.' and i > 0 and i < len(inp) - 1 and inp[i - 1].isdigit() and inp[i + 1].isdigit(): items.append(char) else: items.append(char) mergeitems.append("".join(items)) items = [] else: items.append(char) if items: mergeitems.append("".join(items)) opt = [item for item in mergeitems if not set(item).issubset(punds)] return "\n".join(opt) class TextDictNode: @classmethod def INPUT_TYPES(s): return { "required": { "text": ("STRING", {"multiline": True, "dynamicPrompts": True}), "language": (list(dict_language.keys()),) } } RETURN_TYPES = ("TEXTDICT",) FUNCTION = "encode" CATEGORY = "AIFSH_GPT-SoVITS" def encode(self,text,language): res = { "text": text, "language":language } return (res, ) prompt_sr = 32000 import torchaudio from feature_extractor import cnhubert from transformers import AutoModelForMaskedLM, AutoTokenizer class GSVTTSNode: def __init__(self): self.GPT_weight = None self.SoVITS_weight = None @classmethod def INPUT_TYPES(s): SoVITS_names, GPT_names = get_weights_names(GPT_weight_root, SoVITS_weight_root) return { "required":{ "text_dict": ("TEXTDICT",), "prompt_text_dict":("TEXTDICT",), "prompt_audio":("AUDIO",), "config":("CONFIG",), "GPT_weight":(GPT_names,), "SoVITS_weight":(SoVITS_names,), "how_to_cut":([i18n("不切"), i18n("凑四句一切"), i18n("凑50字一切"), i18n("按中文句号。切"), i18n("按英文句号.切"), i18n("按标点符号切"), ],{ "default": i18n("凑四句一切") }), "speed":("FLOAT",{ "min": 0.6, "max":1.65, "step":0.05, "rond": 0.001, "display":"slider", "default": 1.0 }), "top_k":("INT",{ "min": 1, "max":100, "step":1, "display":"slider", "default": 15 }), "top_p":("FLOAT",{ "min": 0., "max":1., "step":0.05, "rond": 0.001, "display":"slider", "default": 1.0 }), "temperature":("FLOAT",{ "min": 0., "max":1., "step":0.05, "rond": 0.001, "display":"slider", "default": 1.0 }), } } RETURN_TYPES = ("AUDIO",) FUNCTION = "tts" CATEGORY = "AIFSH_GPT-SoVITS" def tts(self,text_dict,prompt_text_dict,prompt_audio, config,GPT_weight,SoVITS_weight,how_to_cut, speed,top_k,top_p,temperature): global ssl_model,is_half,tokenizer,bert_model is_half = config['is_half'] if self.GPT_weight is None: cnhubert.cnhubert_base_path = os.path.join(models_dir,"chinese-hubert-base") ssl_model = cnhubert.get_model() if is_half == True: ssl_model = ssl_model.half().to(device) else: ssl_model = ssl_model.to(device) bert_path = os.path.join(models_dir,"chinese-roberta-wwm-ext-large") tokenizer = AutoTokenizer.from_pretrained(bert_path) bert_model = AutoModelForMaskedLM.from_pretrained(bert_path) if is_half == True: bert_model = bert_model.half().to(device) else: bert_model = bert_model.to(device) text = text_dict ['text'] text_language = text_dict['language'] prompt_text = prompt_text_dict['text'] prompt_language = prompt_text_dict['language'] waveform = prompt_audio['waveform'].squeeze(0) source_sr = prompt_audio['sample_rate'] speech = waveform.mean(dim=0,keepdim=True) if source_sr != prompt_sr: speech = torchaudio.transforms.Resample(orig_freq=source_sr, new_freq=prompt_sr)(speech) if self.SoVITS_weight != SoVITS_weight: self.SoVITS_weight = SoVITS_weight SoVITS_weight_path = os.path.join(models_dir,SoVITS_weight) if "s2G" in SoVITS_weight else os.path.join(work_path,SoVITS_weight) change_sovits_weights(SoVITS_weight_path) if self.GPT_weight != GPT_weight: self.GPT_weight = GPT_weight GPT_weight_path = os.path.join(models_dir,GPT_weight) if "epoch=" in GPT_weight else os.path.join(work_path,GPT_weight) change_gpt_weights(GPT_weight_path) res_audio = get_tts_wav(speech.squeeze(0),prompt_text,prompt_language,text, text_language,how_to_cut,top_k,top_p, temperature,speed) res = { "waveform": res_audio.unsqueeze(0), "sample_rate": prompt_sr, } return (res,) import srt import datetime import traceback import folder_paths from tools.slicer2 import Slicer from faster_whisper import WhisperModel from huggingface_hub import snapshot_download input_dir = folder_paths.get_input_directory() class LoadSRT: @classmethod def INPUT_TYPES(s): files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f)) and f.split('.')[-1] in ["srt", "txt"]] return {"required": {"srt": (sorted(files),)}, } CATEGORY = "AIFSH_CosyVoice" RETURN_TYPES = ("SRT",) FUNCTION = "load_srt" def load_srt(self, srt): srt_path = folder_paths.get_annotated_filepath(srt) return (srt_path,) class PreViewSRT: @classmethod def INPUT_TYPES(s): return {"required": {"srt": ("SRT",)}, } CATEGORY = "AIFSH_GPT-SoVITS" RETURN_TYPES = () OUTPUT_NODE = True FUNCTION = "show_srt" def show_srt(self, srt): srt_name = os.path.basename(srt) dir_name = os.path.dirname(srt) dir_name = os.path.basename(dir_name) with open(srt, 'r',encoding="utf-8") as f: srt_content = f.read() return {"ui": {"srt":[srt_content,srt_name,dir_name]}} import ffmpeg def speed_change(input_audio, speed, sr): # 检查输入数据类型和声道数 if input_audio.dtype != np.int16: raise ValueError("输入音频数据类型必须为 np.int16") # 转换为字节流 raw_audio = input_audio.astype(np.int16).tobytes() # 设置 ffmpeg 输入流 input_stream = ffmpeg.input('pipe:', format='s16le', acodec='pcm_s16le', ar=str(sr), ac=1) # 变速处理 output_stream = input_stream.filter('atempo', speed) # 输出流到管道 out, _ = ( output_stream.output('pipe:', format='s16le', acodec='pcm_s16le') .run(input=raw_audio, capture_stdout=True, capture_stderr=True) ) # 将管道输出解码为 NumPy 数组 processed_audio = np.frombuffer(out, np.int16) return processed_audio class TSCY_Node: def __init__(self): self.ifload_model = True @classmethod def INPUT_TYPES(s): return { "required":{ "from_language":(list(dict_language.keys()),), "to_language": (list(dict_language.keys()),), "prompt_audio":("AUDIO",), "translator":(['alibaba', 'apertium', 'argos', 'baidu', 'bing', 'caiyun', 'cloudTranslation', 'deepl', 'elia', 'google', 'hujiang', 'iciba', 'iflytek', 'iflyrec', 'itranslate', 'judic', 'languageWire', 'lingvanex', 'mglip', 'mirai', 'modernMt', 'myMemory', 'niutrans', 'papago', 'qqFanyi', 'qqTranSmart', 'reverso', 'sogou', 'sysTran', 'tilde', 'translateCom', 'translateMe', 'utibet', 'volcEngine', 'yandex', 'yeekit', 'youdao'],{ "default":"sogou" }), "if_algin":("BOOLEAN",{ "default": True }) }, "optional":{ "tts_srt": ("SRT",), } } RETURN_TYPES = ("AUDIO","SRT",) FUNCTION = "tts" CATEGORY = "AIFSH_GPT-SoVITS" def wishper2gsv(self,lang): if lang in "en": lang = i18n("英文") elif lang in "ko": lang = i18n("韩文") elif lang in "ja": lang = i18n("日文") else: lang = i18n("中文") return lang def gsv2translator(self,lang): if "en" in lang: lang = "en" elif "ja" in lang: lang = "ja" elif "ko" in lang: lang = "ko" else: lang = "zh" return lang def tts(self,from_language,to_language,prompt_audio,translator,if_algin,tts_srt=None): global ssl_model,is_half,tokenizer,bert_model waveform = prompt_audio['waveform'].squeeze(0) source_sr = prompt_audio['sample_rate'] speech = waveform.mean(dim=0,keepdim=True) if source_sr != prompt_sr: speech = torchaudio.transforms.Resample(orig_freq=source_sr, new_freq=prompt_sr)(speech) if self.ifload_model: is_half = True cnhubert.cnhubert_base_path = os.path.join(models_dir,"chinese-hubert-base") ssl_model = cnhubert.get_model() if is_half == True: ssl_model = ssl_model.half().to(device) else: ssl_model = ssl_model.to(device) bert_path = os.path.join(models_dir,"chinese-roberta-wwm-ext-large") tokenizer = AutoTokenizer.from_pretrained(bert_path) bert_model = AutoModelForMaskedLM.from_pretrained(bert_path) if is_half == True: bert_model = bert_model.half().to(device) else: bert_model = bert_model.to(device) change_gpt_weights(os.path.join(models_dir,pretrained_gpt_name[0])) change_sovits_weights(os.path.join(models_dir,pretrained_sovits_name[0])) self.ifload_model = False else: print("use cache model") slicer = Slicer( sr= prompt_sr, threshold= -34, min_length= 4000, min_interval= 300, hop_size= 10, max_sil_kept= 500 ) model_path = os.path.join(models_dir,f"faster-whisper-large-v3") snapshot_download(repo_id=f"Systran/faster-whisper-large-v3",local_dir=model_path) try: model = WhisperModel(model_path, device=device, compute_type="float16") except: return print(traceback.format_exc()) tts_audio = [] tr_to_language = self.gsv2translator(dict_language[to_language]) whisper_language = self.gsv2translator(dict_language[from_language]) subs = [] for i, (chunk, start, end) in enumerate(slicer.slice(speech.numpy()[0])): tmp_max = np.abs(chunk).max() if(tmp_max>1):chunk/=tmp_max chunk = (chunk / tmp_max * (0.9 * 0.25)) + (1 - 0.25) * chunk segments, info = model.transcribe( audio = chunk, beam_size = 5, vad_filter = True, vad_parameters = dict(min_silence_duration_ms=700), language = whisper_language) i_prompt_text = '' if i_prompt_text == '': for segment in segments: i_prompt_text += segment.text i_prompt_audio = torch.from_numpy(chunk) print(f"from {whisper_language} \t {i_prompt_text}") if tts_srt is None: import translators as ts i_tts_text = ts.translate_text(query_text=i_prompt_text,from_language=whisper_language, to_language=tr_to_language,translator=translator) else: with open(tts_srt,"r",encoding="utf-8") as f: sub_str = f.read() i_tts_text = list(srt.parse(sub_str))[i].content print(f"to {tr_to_language}\t{i_tts_text}") i_sub = srt.Subtitle(index=i+1,start=datetime.timedelta(seconds=start/prompt_sr), end=datetime.timedelta(seconds=end/prompt_sr),content=i_tts_text) subs.append(i_sub) i_tts_audio = get_tts_wav(i_prompt_audio,i_prompt_text,from_language,i_tts_text,to_language) i_tts_audio = (i_tts_audio.numpy() * 32768).astype(np.int16) if if_algin: ratio = i_tts_audio.shape[-1] / (end-start) i_tts_audio = [speed_change(i_tts_audio,speed=ratio,sr=prompt_sr) / 32768] print(f"change speed {ratio}") tts_audio.append(i_tts_audio) srt_path = os.path.join(output_dir,"tmp.srt") res_audio = torch.Tensor(np.concatenate(tts_audio, 0)).unsqueeze(0) with open(srt_path,"w",encoding="utf-8") as f: f.write(srt.compose(subs)) res = { "waveform": res_audio, "sample_rate": prompt_sr, } return (res,srt_path,)