inference_main.py

import io
import logging
import time
from pathlib import Path

import librosa
import numpy as np
import soundfile

from inference import infer_tool
from inference import slicer
from inference.infer_tool import Svc

logging.getLogger('numba').setLevel(logging.WARNING)
chunks_dict = infer_tool.read_temp("inference/chunks_temp.json")

model_path = "logs/48k/G_700000.pth"
config_path = "configs/suiji.json"
svc_model = Svc(model_path, config_path)
infer_tool.mkdir(["raw", "results"])

# 支持多个wav文件，放在raw文件夹下
clean_names = ["2043_7"]
# 例如：clean_names = ["千千阙歌_1", "千千阙歌_2", "千千阙歌_3", "千千阙歌_4", "千千阙歌_5", "千千阙歌_6", "千千阙歌_7", "千千阙歌_8"]
# 中文字符过多会报编码错误，分两批就好
trans = [0]  # 音高调整，支持正负（半音）
spk_list = ['suiji']  # 每次同时合成多语者音色
slice_db = -40  # 默认-40，嘈杂的音频可以-30，干声保留呼吸可以-50
wav_format = 'wav'  # 音频输出格式

infer_tool.fill_a_to_b(trans, clean_names)
for clean_name, tran in zip(clean_names, trans):
    raw_audio_path = f"raw/{clean_name}"
    if "." not in raw_audio_path:
        raw_audio_path += ".wav"
    infer_tool.format_wav(raw_audio_path)
    wav_path = Path(raw_audio_path).with_suffix('.wav')
    chunks = slicer.cut(wav_path, db_thresh=slice_db)
    audio_data, audio_sr = slicer.chunks2audio(wav_path, chunks)

    for spk in spk_list:
        audio = []
        for (slice_tag, data) in audio_data:
            print(f'#=====segment start, {round(len(data) / audio_sr, 3)}s======')
            length = int(np.ceil(len(data) / audio_sr * svc_model.target_sample))
            raw_path = io.BytesIO()
            soundfile.write(raw_path, data, audio_sr, format="wav")
            raw_path.seek(0)
            if slice_tag:
                print('jump empty segment')
                _audio = np.zeros(length)
            else:
                out_audio, out_sr = svc_model.infer(spk, tran, raw_path)
                _audio = out_audio.cpu().numpy()
            audio.extend(list(_audio))

        res_path = f'./results/{clean_name}_{tran}key_{spk}.{wav_format}'
        soundfile.write(res_path, audio, svc_model.target_sample, format=wav_format)