import sys, os if sys.platform == "darwin": os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1" import logging logging.getLogger("numba").setLevel(logging.WARNING) logging.getLogger("markdown_it").setLevel(logging.WARNING) logging.getLogger("urllib3").setLevel(logging.WARNING) logging.getLogger("matplotlib").setLevel(logging.WARNING) logging.basicConfig(level=logging.INFO, format="| %(name)s | %(levelname)s | %(message)s") logger = logging.getLogger(__name__) import torch import argparse import commons import utils from models import SynthesizerTrn from text.symbols import symbols from text import cleaned_text_to_sequence, get_bert from text.cleaner import clean_text import gradio as gr import webbrowser net_g = None def get_text(text, language_str, hps): norm_text, phone, tone, word2ph = clean_text(text, language_str) phone, tone, language = cleaned_text_to_sequence(phone, tone, language_str) if hps.data.add_blank: phone = commons.intersperse(phone, 0) tone = commons.intersperse(tone, 0) language = commons.intersperse(language, 0) for i in range(len(word2ph)): word2ph[i] = word2ph[i] * 2 word2ph[0] += 1 bert = get_bert(norm_text, word2ph, language_str) del word2ph assert bert.shape[-1] == len(phone) phone = torch.LongTensor(phone) tone = torch.LongTensor(tone) language = torch.LongTensor(language) return bert, phone, tone, language import soundfile as sf def infer(text, sdp_ratio, noise_scale, noise_scale_w, length_scale, sid): global net_g bert, phones, tones, lang_ids = get_text(text, "ZH", hps) with torch.no_grad(): x_tst=phones.to(device).unsqueeze(0) tones=tones.to(device).unsqueeze(0) lang_ids=lang_ids.to(device).unsqueeze(0) bert = bert.to(device).unsqueeze(0) x_tst_lengths = torch.LongTensor([phones.size(0)]).to(device) del phones speakers = torch.LongTensor([hps.data.spk2id[sid]]).to(device) audio = net_g.infer(x_tst, x_tst_lengths, speakers, tones, lang_ids, bert, sdp_ratio=sdp_ratio , noise_scale=noise_scale, noise_scale_w=noise_scale_w, length_scale=length_scale)[0][0,0].data.cpu().float().numpy() del x_tst, tones, lang_ids, bert, x_tst_lengths, speakers sf.write("tmp.wav", audio, 44100) return audio def convert_wav_to_ogg(wav_file): os.makedirs('out', exist_ok=True) filename = os.path.splitext(os.path.basename(wav_file.name))[0] output_path_ogg = os.path.join('out', f"out.ogg") renamed_input_path = os.path.join('in', f"in.wav") os.makedirs('in', exist_ok=True) os.rename(wav_file.name, renamed_input_path) command = ["ffmpeg", "-i", renamed_input_path, "-acodec", "libopus", "-y", output_path_ogg] os.system(" ".join(command)) return output_path_ogg def tts_fn(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale): with torch.no_grad(): audio = infer(text, sdp_ratio=sdp_ratio, noise_scale=noise_scale, noise_scale_w=noise_scale_w, length_scale=length_scale, sid=speaker) with open('tmp.wav', 'rb') as wav_file: newogg = convert_wav_to_ogg(wav_file) return "Success", (hps.data.sampling_rate, audio),newogg if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument("--model_dir", default="./logs/kh/kh.pth", help="path of your model") parser.add_argument("--config_dir", default="./configs/config.json", help="path of your config file") parser.add_argument("--share", default=False, help="make link public") parser.add_argument("-d", "--debug", action="store_true", help="enable DEBUG-LEVEL log") args = parser.parse_args() if args.debug: logger.info("Enable DEBUG-LEVEL log") logging.basicConfig(level=logging.DEBUG) hps = utils.get_hparams_from_file(args.config_dir) device = "cuda:0" if torch.cuda.is_available() else "cpu" ''' device = ( "cuda:0" if torch.cuda.is_available() else ( "mps" if sys.platform == "darwin" and torch.backends.mps.is_available() else "cpu" ) ) ''' net_g = SynthesizerTrn( len(symbols), hps.data.filter_length // 2 + 1, hps.train.segment_size // hps.data.hop_length, n_speakers=hps.data.n_speakers, **hps.model).to(device) _ = net_g.eval() _ = utils.load_checkpoint(args.model_dir, net_g, None, skip_optimizer=True) speaker_ids = hps.data.spk2id speakers = list(speaker_ids.keys()) with gr.Blocks() as app: with gr.Row(): with gr.Column(): gr.Markdown(value=""" 男声朗读 Bert-Vits2在线语音生成\n 1、模型作者：数字星瞳企划 https://t.me/xingtong25680 \n 2、原项目地址：https://github.com/Stardust-minus/Bert-VITS2\n 3、使用此模型进行二创请注明AI生成，以及原项目地址\n 4、素材来自散文朗读比赛，严禁将此项目用于一切违反《中华人民共和国宪法》，《中华人民共和国刑法》，《中华人民共和国治安管理处罚法》和《中华人民共和国民法典》之用途。严禁用于任何政治相关用途。 \n """) text = gr.TextArea(label="Text", placeholder="Input Text Here", value="这里是数字星瞳企画，请在电报搜索星瞳全拼加二五六八零，获取最新更新进展。") speaker = gr.Dropdown(choices=speakers, value=speakers[0], label='Speaker') sdp_ratio = gr.Slider(minimum=0, maximum=1, value=0.2, step=0.01, label='语调变化') noise_scale = gr.Slider(minimum=0.1, maximum=1.5, value=0.6, step=0.01, label='感情变化') noise_scale_w = gr.Slider(minimum=0.1, maximum=1.4, value=0.8, step=0.01, label='音节发音长度变化') length_scale = gr.Slider(minimum=0.1, maximum=2, value=1, step=0.01, label='语速') btn = gr.Button("开启AI语音之旅吧!", variant="primary") with gr.Column(): text_output = gr.Textbox(label="Message") audio_output = gr.Audio(label="Output Audio") ogg_output = gr.File(label="Converted OGG file") gr.Markdown(value=""" 模型汇总：\n 星瞳整合 https://huggingface.co/spaces/digitalxingtong/Xingtong-All-in-One\n 男声朗读 https://huggingface.co/spaces/digitalxingtong/Kanghui-Read-Bert-VITS2 \n 男声朗读（长文本） https://huggingface.co/spaces/digitalxingtong/Kanghui-Longread-Bert-VITS2\n """) btn.click(tts_fn, inputs=[text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale], outputs=[text_output, audio_output,ogg_output]) app.launch(show_error=True)