# flake8: noqa: E402 import sys, os import logging import os import time import numpy as np # 假设你使用NumPy来处理音频数据 import shutil # 用于删除文件夹和文件 from scipy.io import wavfile 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 import numpy as np net_g = None if sys.platform == "darwin" and torch.backends.mps.is_available(): device = "mps" os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1" else: device = "cuda" 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, device) del word2ph assert bert.shape[-1] == len(phone), phone if language_str == "ZH": bert = bert ja_bert = torch.zeros(768, len(phone)) elif language_str == "JP": ja_bert = bert bert = torch.zeros(1024, len(phone)) else: bert = torch.zeros(1024, len(phone)) ja_bert = torch.zeros(768, len(phone)) assert bert.shape[-1] == len( phone ), f"Bert seq len {bert.shape[-1]} != {len(phone)}" phone = torch.LongTensor(phone) tone = torch.LongTensor(tone) language = torch.LongTensor(language) return bert, ja_bert, phone, tone, language def infer(text, sdp_ratio, noise_scale, noise_scale_w, length_scale, sid, language): global net_g bert, ja_bert, phones, tones, lang_ids = get_text(text, language, 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) ja_bert = ja_bert.to(device).unsqueeze(0) x_tst_lengths = torch.LongTensor([phones.size(0)]).to(device) #print(x_tst.type(), tones.type(), lang_ids.type(), bert.type(), ja_bert.type(), x_tst_lengths.type()) 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, ja_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 torch.cuda.empty_cache() return audio __LOG__ = "./generation_logs.txt" def tts_fn(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale, language): # 清空 ./infer_save 文件夹 if os.path.exists('./infer_save'): shutil.rmtree('./infer_save') os.makedirs('./infer_save') slices = text.split("\n") slices = [slice for slice in slices if slice.strip() != ""] audio_list = [] with torch.no_grad(): with open(__LOG__,"a",encoding="UTF-8") as f: for slice in slices: assert len(slice) < 150 # 限制输入的文本长度 audio = infer(slice, sdp_ratio=sdp_ratio, noise_scale=noise_scale, noise_scale_w=noise_scale_w, length_scale=length_scale, sid=speaker, language=language) audio_list.append(audio) # 创建唯一的文件名 timestamp = str(int(time.time() * 1000)) audio_file_path = f'./infer_save/audio_{timestamp}.wav' # 保存音频数据到.wav文件 wavfile.write(audio_file_path, hps.data.sampling_rate, audio) silence = np.zeros(hps.data.sampling_rate, dtype=np.int16) # 生成1秒的静音 audio_list.append(silence) # 将静音添加到列表中 f.write(f"{slice} | {speaker}\n") print(f"{slice} | {speaker}") audio_concat = np.concatenate(audio_list) return "Success", (hps.data.sampling_rate, audio_concat) if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument( "-m", "--model", default="./logs/umamusume/G_300500.pth", help="path of your model" ) parser.add_argument( "-c", "--config", default="./configs/config.json", help="path of your config file", ) parser.add_argument( "--share", default=False, help="make link public", action="store_true" ) 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) 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() # _ = net_g.half() # _ = net_g.to(dtype=torch.bfloat16) _ = utils.load_checkpoint(args.model, net_g, None, skip_optimizer=True) speaker_ids = hps.data.spk2id speakers = list(speaker_ids.keys()) languages = ["ZH", "JP"] with gr.Blocks() as app: with gr.Row(): with gr.Column(): text = gr.TextArea( label="Text", placeholder="Input Text Here", value="はりきっていこう！", ) speaker = gr.Dropdown( choices=speakers, value=speakers[0], label="Speaker" ) sdp_ratio = gr.Slider( minimum=0, maximum=1, value=0.2, step=0.1, label="SDP Ratio" ) noise_scale = gr.Slider( minimum=0.1, maximum=2, value=0.6, step=0.1, label="Noise Scale" ) noise_scale_w = gr.Slider( minimum=0.1, maximum=2, value=0.8, step=0.1, label="Noise Scale W" ) length_scale = gr.Slider( minimum=0.1, maximum=2, value=1, step=0.1, label="Length Scale" ) language = gr.Dropdown( choices=languages, value=languages[1], label="Language" ) btn = gr.Button("Generate!", variant="primary") with gr.Column(): text_output = gr.Textbox(label="Message") audio_output = gr.Audio(label="Output Audio") gr.Markdown("# 赛马娘 Bert-VITS2 语音合成\n" "Project page：[GitHub](https://github.com/fishaudio/Bert-VITS2)\n" "- 本项目在日语方面有所欠缺，特别是音调的设计上，需要帮助。\n" "- このプロジェクトは、日本語の方面で不足しています。特に、音調の設計に関して助けが欲しいです。") btn.click( tts_fn, inputs=[ text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale, language, ], outputs=[text_output, audio_output], ) app.launch()