From 72a18e66b6317e6c67fde52f00452e9abc271b88 Mon Sep 17 00:00:00 2001 From: RVC-Boss <129054828+RVC-Boss@users.noreply.github.com> Date: Wed, 13 Sep 2023 11:33:48 +0800 Subject: [PATCH] Add files via upload --- modules.py | 308 +++++++++++++++++++++++++++++++++++ pipeline.py | 457 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 765 insertions(+) create mode 100644 modules.py create mode 100644 pipeline.py diff --git a/modules.py b/modules.py new file mode 100644 index 0000000..81a21a6 --- /dev/null +++ b/modules.py @@ -0,0 +1,308 @@ +import traceback +import logging + +logger = logging.getLogger(__name__) + +import numpy as np +import soundfile as sf +import torch +from io import BytesIO + +from infer.lib.audio import load_audio, wav2 +from infer.lib.infer_pack.models import ( + SynthesizerTrnMs256NSFsid, + SynthesizerTrnMs256NSFsid_nono, + SynthesizerTrnMs768NSFsid, + SynthesizerTrnMs768NSFsid_nono, +) +from infer.modules.vc.pipeline import Pipeline +from infer.modules.vc.utils import * + + +class VC: + def __init__(self, config): + self.n_spk = None + self.tgt_sr = None + self.net_g = None + self.pipeline = None + self.cpt = None + self.version = None + self.if_f0 = None + self.version = None + self.hubert_model = None + + self.config = config + + def get_vc(self, sid, *to_return_protect): + logger.info("Get sid: " + sid) + + to_return_protect0 = { + "visible": self.if_f0 != 0, + "value": to_return_protect[0] + if self.if_f0 != 0 and to_return_protect + else 0.5, + "__type__": "update", + } + to_return_protect1 = { + "visible": self.if_f0 != 0, + "value": to_return_protect[1] + if self.if_f0 != 0 and to_return_protect + else 0.33, + "__type__": "update", + } + + if sid == "" or sid == []: + if self.hubert_model is not None: # 考虑到轮询, 需要加个判断看是否 sid 是由有模型切换到无模型的 + logger.info("Clean model cache") + del ( + self.net_g, + self.n_spk, + self.vc, + self.hubert_model, + self.tgt_sr, + ) # ,cpt + self.hubert_model = ( + self.net_g + ) = self.n_spk = self.vc = self.hubert_model = self.tgt_sr = None + if torch.cuda.is_available(): + torch.cuda.empty_cache() + ###楼下不这么折腾清理不干净 + self.if_f0 = self.cpt.get("f0", 1) + self.version = self.cpt.get("version", "v1") + if self.version == "v1": + if self.if_f0 == 1: + self.net_g = SynthesizerTrnMs256NSFsid( + *self.cpt["config"], is_half=self.config.is_half + ) + else: + self.net_g = SynthesizerTrnMs256NSFsid_nono(*self.cpt["config"]) + elif self.version == "v2": + if self.if_f0 == 1: + self.net_g = SynthesizerTrnMs768NSFsid( + *self.cpt["config"], is_half=self.config.is_half + ) + else: + self.net_g = SynthesizerTrnMs768NSFsid_nono(*self.cpt["config"]) + del self.net_g, self.cpt + if torch.cuda.is_available(): + torch.cuda.empty_cache() + return ( + {"visible": False, "__type__": "update"}, + { + "visible": True, + "value": to_return_protect0, + "__type__": "update", + }, + { + "visible": True, + "value": to_return_protect1, + "__type__": "update", + }, + "", + "", + ) + person = f'{os.getenv("weight_root")}/{sid}' + logger.info(f"Loading: {person}") + + self.cpt = torch.load(person, map_location="cpu") + self.tgt_sr = self.cpt["config"][-1] + self.cpt["config"][-3] = self.cpt["weight"]["emb_g.weight"].shape[0] # n_spk + self.if_f0 = self.cpt.get("f0", 1) + self.version = self.cpt.get("version", "v1") + + synthesizer_class = { + ("v1", 1): SynthesizerTrnMs256NSFsid, + ("v1", 0): SynthesizerTrnMs256NSFsid_nono, + ("v2", 1): SynthesizerTrnMs768NSFsid, + ("v2", 0): SynthesizerTrnMs768NSFsid_nono, + } + + self.net_g = synthesizer_class.get( + (self.version, self.if_f0), SynthesizerTrnMs256NSFsid + )(*self.cpt["config"], is_half=self.config.is_half) + + del self.net_g.enc_q + + self.net_g.load_state_dict(self.cpt["weight"], strict=False) + self.net_g.eval().to(self.config.device) + if self.config.is_half: + self.net_g = self.net_g.half() + else: + self.net_g = self.net_g.float() + + self.pipeline = Pipeline(self.tgt_sr, self.config) + n_spk = self.cpt["config"][-3] + index = {"value": get_index_path_from_model(sid), "__type__": "update"} + logger.info("Select index: " + index["value"]) + + return ( + ( + {"visible": True, "maximum": n_spk, "__type__": "update"}, + to_return_protect0, + to_return_protect1, + index, + index, + ) + if to_return_protect + else {"visible": True, "maximum": n_spk, "__type__": "update"} + ) + + def vc_single( + self, + sid, + input_audio_path, + f0_up_key, + f0_file, + f0_method, + file_index, + file_index2, + index_rate, + filter_radius, + resample_sr, + rms_mix_rate, + protect, + ): + if input_audio_path is None: + return "You need to upload an audio", None + f0_up_key = int(f0_up_key) + try: + audio = load_audio(input_audio_path, 16000) + audio_max = np.abs(audio).max() / 0.95 + if audio_max > 1: + audio /= audio_max + times = [0, 0, 0] + + if self.hubert_model is None: + self.hubert_model = load_hubert(self.config) + + file_index = ( + ( + file_index.strip(" ") + .strip('"') + .strip("\n") + .strip('"') + .strip(" ") + .replace("trained", "added") + ) + if file_index != "" + else file_index2 + ) # 防止小白写错，自动帮他替换掉 + + audio_opt = self.pipeline.pipeline( + self.hubert_model, + self.net_g, + sid, + audio, + input_audio_path, + times, + f0_up_key, + f0_method, + file_index, + index_rate, + self.if_f0, + filter_radius, + self.tgt_sr, + resample_sr, + rms_mix_rate, + self.version, + protect, + f0_file, + ) + if self.tgt_sr != resample_sr >= 16000: + tgt_sr = resample_sr + else: + tgt_sr = self.tgt_sr + index_info = ( + "Index:\n%s." % file_index + if os.path.exists(file_index) + else "Index not used." + ) + return ( + "Success.\n%s\nTime:\nnpy: %.2fs, f0: %.2fs, infer: %.2fs." + % (index_info, *times), + (tgt_sr, audio_opt), + ) + except: + info = traceback.format_exc() + logger.warn(info) + return info, (None, None) + + def vc_multi( + self, + sid, + dir_path, + opt_root, + paths, + f0_up_key, + f0_method, + file_index, + file_index2, + index_rate, + filter_radius, + resample_sr, + rms_mix_rate, + protect, + format1, + ): + try: + dir_path = ( + dir_path.strip(" ").strip('"').strip("\n").strip('"').strip(" ") + ) # 防止小白拷路径头尾带了空格和"和回车 + opt_root = opt_root.strip(" ").strip('"').strip("\n").strip('"').strip(" ") + os.makedirs(opt_root, exist_ok=True) + try: + if dir_path != "": + paths = [ + os.path.join(dir_path, name) for name in os.listdir(dir_path) + ] + else: + paths = [path.name for path in paths] + except: + traceback.print_exc() + paths = [path.name for path in paths] + infos = [] + for path in paths: + info, opt = self.vc_single( + sid, + path, + f0_up_key, + None, + f0_method, + file_index, + file_index2, + # file_big_npy, + index_rate, + filter_radius, + resample_sr, + rms_mix_rate, + protect, + ) + if "Success" in info: + try: + tgt_sr, audio_opt = opt + if format1 in ["wav", "flac"]: + sf.write( + "%s/%s.%s" + % (opt_root, os.path.basename(path), format1), + audio_opt, + tgt_sr, + ) + else: + path = "%s/%s.%s" % (opt_root, os.path.basename(path), format1) + with BytesIO() as wavf: + sf.write( + wavf, + audio_opt, + tgt_sr, + format="wav" + ) + wavf.seek(0, 0) + with open(path, "wb") as outf: + wav2(wavf, outf, format1) + except: + info += traceback.format_exc() + infos.append("%s->%s" % (os.path.basename(path), info)) + yield "\n".join(infos) + yield "\n".join(infos) + except: + yield traceback.format_exc() diff --git a/pipeline.py b/pipeline.py new file mode 100644 index 0000000..0c22584 --- /dev/null +++ b/pipeline.py @@ -0,0 +1,457 @@ +import os +import sys +import traceback +import logging + +logger = logging.getLogger(__name__) + +from functools import lru_cache +from time import time as ttime + +import faiss +import librosa +import numpy as np +import parselmouth +import pyworld +import torch +import torch.nn.functional as F +import torchcrepe +from scipy import signal + +now_dir = os.getcwd() +sys.path.append(now_dir) + +bh, ah = signal.butter(N=5, Wn=48, btype="high", fs=16000) + +input_audio_path2wav = {} + + +@lru_cache +def cache_harvest_f0(input_audio_path, fs, f0max, f0min, frame_period): + audio = input_audio_path2wav[input_audio_path] + f0, t = pyworld.harvest( + audio, + fs=fs, + f0_ceil=f0max, + f0_floor=f0min, + frame_period=frame_period, + ) + f0 = pyworld.stonemask(audio, f0, t, fs) + return f0 + + +def change_rms(data1, sr1, data2, sr2, rate): # 1是输入音频，2是输出音频,rate是2的占比 + # print(data1.max(),data2.max()) + rms1 = librosa.feature.rms( + y=data1, frame_length=sr1 // 2 * 2, hop_length=sr1 // 2 + ) # 每半秒一个点 + rms2 = librosa.feature.rms(y=data2, frame_length=sr2 // 2 * 2, hop_length=sr2 // 2) + rms1 = torch.from_numpy(rms1) + rms1 = F.interpolate( + rms1.unsqueeze(0), size=data2.shape[0], mode="linear" + ).squeeze() + rms2 = torch.from_numpy(rms2) + rms2 = F.interpolate( + rms2.unsqueeze(0), size=data2.shape[0], mode="linear" + ).squeeze() + rms2 = torch.max(rms2, torch.zeros_like(rms2) + 1e-6) + data2 *= ( + torch.pow(rms1, torch.tensor(1 - rate)) + * torch.pow(rms2, torch.tensor(rate - 1)) + ).numpy() + return data2 + + +class Pipeline(object): + def __init__(self, tgt_sr, config): + self.x_pad, self.x_query, self.x_center, self.x_max, self.is_half = ( + config.x_pad, + config.x_query, + config.x_center, + config.x_max, + config.is_half, + ) + self.sr = 16000 # hubert输入采样率 + self.window = 160 # 每帧点数 + self.t_pad = self.sr * self.x_pad # 每条前后pad时间 + self.t_pad_tgt = tgt_sr * self.x_pad + self.t_pad2 = self.t_pad * 2 + self.t_query = self.sr * self.x_query # 查询切点前后查询时间 + self.t_center = self.sr * self.x_center # 查询切点位置 + self.t_max = self.sr * self.x_max # 免查询时长阈值 + self.device = config.device + + def get_f0( + self, + input_audio_path, + x, + p_len, + f0_up_key, + f0_method, + filter_radius, + inp_f0=None, + ): + global input_audio_path2wav + time_step = self.window / self.sr * 1000 + f0_min = 50 + f0_max = 1100 + f0_mel_min = 1127 * np.log(1 + f0_min / 700) + f0_mel_max = 1127 * np.log(1 + f0_max / 700) + if f0_method == "pm": + f0 = ( + parselmouth.Sound(x, self.sr) + .to_pitch_ac( + time_step=time_step / 1000, + voicing_threshold=0.6, + pitch_floor=f0_min, + pitch_ceiling=f0_max, + ) + .selected_array["frequency"] + ) + pad_size = (p_len - len(f0) + 1) // 2 + if pad_size > 0 or p_len - len(f0) - pad_size > 0: + f0 = np.pad( + f0, [[pad_size, p_len - len(f0) - pad_size]], mode="constant" + ) + elif f0_method == "harvest": + input_audio_path2wav[input_audio_path] = x.astype(np.double) + f0 = cache_harvest_f0(input_audio_path, self.sr, f0_max, f0_min, 10) + if filter_radius > 2: + f0 = signal.medfilt(f0, 3) + elif f0_method == "crepe": + model = "full" + # Pick a batch size that doesn't cause memory errors on your gpu + batch_size = 512 + # Compute pitch using first gpu + audio = torch.tensor(np.copy(x))[None].float() + f0, pd = torchcrepe.predict( + audio, + self.sr, + self.window, + f0_min, + f0_max, + model, + batch_size=batch_size, + device=self.device, + return_periodicity=True, + ) + pd = torchcrepe.filter.median(pd, 3) + f0 = torchcrepe.filter.mean(f0, 3) + f0[pd < 0.1] = 0 + f0 = f0[0].cpu().numpy() + elif f0_method == "rmvpe": + if not hasattr(self, "model_rmvpe"): + from infer.lib.rmvpe import RMVPE + + logger.info( + "Loading rmvpe model,%s" % "%s/rmvpe.pt" % os.environ["rmvpe_root"] + ) + self.model_rmvpe = RMVPE( + "%s/rmvpe.pt" % os.environ["rmvpe_root"], + is_half=self.is_half, + device=self.device, + ) + f0 = self.model_rmvpe.infer_from_audio(x, thred=0.03) + + if "privateuseone" in str(self.device): # clean ortruntime memory + del self.model_rmvpe.model + del self.model_rmvpe + logger.info("Cleaning ortruntime memory") + + f0 *= pow(2, f0_up_key / 12) + # with open("test.txt","w")as f:f.write("\n".join([str(i)for i in f0.tolist()])) + tf0 = self.sr // self.window # 每秒f0点数 + if inp_f0 is not None: + delta_t = np.round( + (inp_f0[:, 0].max() - inp_f0[:, 0].min()) * tf0 + 1 + ).astype("int16") + replace_f0 = np.interp( + list(range(delta_t)), inp_f0[:, 0] * 100, inp_f0[:, 1] + ) + shape = f0[self.x_pad * tf0 : self.x_pad * tf0 + len(replace_f0)].shape[0] + f0[self.x_pad * tf0 : self.x_pad * tf0 + len(replace_f0)] = replace_f0[ + :shape + ] + # with open("test_opt.txt","w")as f:f.write("\n".join([str(i)for i in f0.tolist()])) + f0bak = f0.copy() + f0_mel = 1127 * np.log(1 + f0 / 700) + f0_mel[f0_mel > 0] = (f0_mel[f0_mel > 0] - f0_mel_min) * 254 / ( + f0_mel_max - f0_mel_min + ) + 1 + f0_mel[f0_mel <= 1] = 1 + f0_mel[f0_mel > 255] = 255 + f0_coarse = np.rint(f0_mel).astype(np.int32) + return f0_coarse, f0bak # 1-0 + + def vc( + self, + model, + net_g, + sid, + audio0, + pitch, + pitchf, + times, + index, + big_npy, + index_rate, + version, + protect, + ): # ,file_index,file_big_npy + feats = torch.from_numpy(audio0) + if self.is_half: + feats = feats.half() + else: + feats = feats.float() + if feats.dim() == 2: # double channels + feats = feats.mean(-1) + assert feats.dim() == 1, feats.dim() + feats = feats.view(1, -1) + padding_mask = torch.BoolTensor(feats.shape).to(self.device).fill_(False) + + inputs = { + "source": feats.to(self.device), + "padding_mask": padding_mask, + "output_layer": 9 if version == "v1" else 12, + } + t0 = ttime() + with torch.no_grad(): + logits = model.extract_features(**inputs) + feats = model.final_proj(logits[0]) if version == "v1" else logits[0] + if protect < 0.5 and pitch is not None and pitchf is not None: + feats0 = feats.clone() + if ( + not isinstance(index, type(None)) + and not isinstance(big_npy, type(None)) + and index_rate != 0 + ): + npy = feats[0].cpu().numpy() + if self.is_half: + npy = npy.astype("float32") + + # _, I = index.search(npy, 1) + # npy = big_npy[I.squeeze()] + + score, ix = index.search(npy, k=8) + weight = np.square(1 / score) + weight /= weight.sum(axis=1, keepdims=True) + npy = np.sum(big_npy[ix] * np.expand_dims(weight, axis=2), axis=1) + + if self.is_half: + npy = npy.astype("float16") + feats = ( + torch.from_numpy(npy).unsqueeze(0).to(self.device) * index_rate + + (1 - index_rate) * feats + ) + + feats = F.interpolate(feats.permute(0, 2, 1), scale_factor=2).permute(0, 2, 1) + if protect < 0.5 and pitch is not None and pitchf is not None: + feats0 = F.interpolate(feats0.permute(0, 2, 1), scale_factor=2).permute( + 0, 2, 1 + ) + t1 = ttime() + p_len = audio0.shape[0] // self.window + if feats.shape[1] < p_len: + p_len = feats.shape[1] + if pitch is not None and pitchf is not None: + pitch = pitch[:, :p_len] + pitchf = pitchf[:, :p_len] + + if protect < 0.5 and pitch is not None and pitchf is not None: + pitchff = pitchf.clone() + pitchff[pitchf > 0] = 1 + pitchff[pitchf < 1] = protect + pitchff = pitchff.unsqueeze(-1) + feats = feats * pitchff + feats0 * (1 - pitchff) + feats = feats.to(feats0.dtype) + p_len = torch.tensor([p_len], device=self.device).long() + with torch.no_grad(): + hasp = pitch is not None and pitchf is not None + arg = (feats, p_len, pitch, pitchf, sid) if hasp else (feats, p_len, sid) + audio1 = (net_g.infer(*arg)[0][0, 0]).data.cpu().float().numpy() + del hasp, arg + del feats, p_len, padding_mask + if torch.cuda.is_available(): + torch.cuda.empty_cache() + t2 = ttime() + times[0] += t1 - t0 + times[2] += t2 - t1 + return audio1 + + def pipeline( + self, + model, + net_g, + sid, + audio, + input_audio_path, + times, + f0_up_key, + f0_method, + file_index, + index_rate, + if_f0, + filter_radius, + tgt_sr, + resample_sr, + rms_mix_rate, + version, + protect, + f0_file=None, + ): + if ( + file_index != "" + # and file_big_npy != "" + # and os.path.exists(file_big_npy) == True + and os.path.exists(file_index) + and index_rate != 0 + ): + try: + index = faiss.read_index(file_index) + # big_npy = np.load(file_big_npy) + big_npy = index.reconstruct_n(0, index.ntotal) + except: + traceback.print_exc() + index = big_npy = None + else: + index = big_npy = None + audio = signal.filtfilt(bh, ah, audio) + audio_pad = np.pad(audio, (self.window // 2, self.window // 2), mode="reflect") + opt_ts = [] + if audio_pad.shape[0] > self.t_max: + audio_sum = np.zeros_like(audio) + for i in range(self.window): + audio_sum += audio_pad[i : i - self.window] + for t in range(self.t_center, audio.shape[0], self.t_center): + opt_ts.append( + t + - self.t_query + + np.where( + np.abs(audio_sum[t - self.t_query : t + self.t_query]) + == np.abs(audio_sum[t - self.t_query : t + self.t_query]).min() + )[0][0] + ) + s = 0 + audio_opt = [] + t = None + t1 = ttime() + audio_pad = np.pad(audio, (self.t_pad, self.t_pad), mode="reflect") + p_len = audio_pad.shape[0] // self.window + inp_f0 = None + if hasattr(f0_file, "name"): + try: + with open(f0_file.name, "r") as f: + lines = f.read().strip("\n").split("\n") + inp_f0 = [] + for line in lines: + inp_f0.append([float(i) for i in line.split(",")]) + inp_f0 = np.array(inp_f0, dtype="float32") + except: + traceback.print_exc() + sid = torch.tensor(sid, device=self.device).unsqueeze(0).long() + pitch, pitchf = None, None + if if_f0 == 1: + pitch, pitchf = self.get_f0( + input_audio_path, + audio_pad, + p_len, + f0_up_key, + f0_method, + filter_radius, + inp_f0, + ) + pitch = pitch[:p_len] + pitchf = pitchf[:p_len] + if "mps" not in str(self.device) or "xpu" not in str(self.device): + pitchf = pitchf.astype(np.float32) + pitch = torch.tensor(pitch, device=self.device).unsqueeze(0).long() + pitchf = torch.tensor(pitchf, device=self.device).unsqueeze(0).float() + t2 = ttime() + times[1] += t2 - t1 + for t in opt_ts: + t = t // self.window * self.window + if if_f0 == 1: + audio_opt.append( + self.vc( + model, + net_g, + sid, + audio_pad[s : t + self.t_pad2 + self.window], + pitch[:, s // self.window : (t + self.t_pad2) // self.window], + pitchf[:, s // self.window : (t + self.t_pad2) // self.window], + times, + index, + big_npy, + index_rate, + version, + protect, + )[self.t_pad_tgt : -self.t_pad_tgt] + ) + else: + audio_opt.append( + self.vc( + model, + net_g, + sid, + audio_pad[s : t + self.t_pad2 + self.window], + None, + None, + times, + index, + big_npy, + index_rate, + version, + protect, + )[self.t_pad_tgt : -self.t_pad_tgt] + ) + s = t + if if_f0 == 1: + audio_opt.append( + self.vc( + model, + net_g, + sid, + audio_pad[t:], + pitch[:, t // self.window :] if t is not None else pitch, + pitchf[:, t // self.window :] if t is not None else pitchf, + times, + index, + big_npy, + index_rate, + version, + protect, + )[self.t_pad_tgt : -self.t_pad_tgt] + ) + else: + audio_opt.append( + self.vc( + model, + net_g, + sid, + audio_pad[t:], + None, + None, + times, + index, + big_npy, + index_rate, + version, + protect, + )[self.t_pad_tgt : -self.t_pad_tgt] + ) + audio_opt = np.concatenate(audio_opt) + if rms_mix_rate != 1: + audio_opt = change_rms(audio, 16000, audio_opt, tgt_sr, rms_mix_rate) + if tgt_sr != resample_sr >= 16000: + audio_opt = librosa.resample( + audio_opt, orig_sr=tgt_sr, target_sr=resample_sr + ) + audio_max = np.abs(audio_opt).max() / 0.99 + max_int16 = 32768 + if audio_max > 1: + max_int16 /= audio_max + audio_opt = (audio_opt * max_int16).astype(np.int16) + del pitch, pitchf, sid + if torch.cuda.is_available(): + torch.cuda.empty_cache() + return audio_opt