From 8658f9aac5213d0ce678e99543fc65100a608d4b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stardust=C2=B7=E5=87=8F?= <2225664821@qq.com> Date: Thu, 3 Aug 2023 18:53:01 +0800 Subject: [PATCH] Add files via upload --- server.py | 98 +++++++++++++++++++++++++++++++------------------------ 1 file changed, 55 insertions(+), 43 deletions(-) diff --git a/server.py b/server.py index db14cbb..30d926f 100644 --- a/server.py +++ b/server.py @@ -17,6 +17,8 @@ import utils from models import SynthesizerTrn from text.symbols import symbols from text import text_to_sequence +from text import cleaned_text_to_sequence,_symbol_to_id, get_bert +from text.cleaner import clean_text from scipy.io import wavfile # Get ffmpeg path @@ -25,28 +27,60 @@ ffmpeg_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), "ffmpeg") # Flask Init app = Flask(__name__) app.config['JSON_AS_ASCII'] = False -# Text Preprocess -def get_text(text, hps): - text_norm = text_to_sequence(text, hps.data.text_cleaners) +def get_text(text, language_str, hps): + norm_text, phone, tone, word2ph = clean_text(text, language_str) + print([f"{p}{t}" for p, t in zip(phone, tone)]) + phone, tone, language = cleaned_text_to_sequence(phone, tone, language_str) + if hps.data.add_blank: - text_norm = commons.intersperse(text_norm, 0) - text_norm = torch.LongTensor(text_norm) - return text_norm + 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) + + assert bert.shape[-1] == len(phone) + + phone = torch.LongTensor(phone) + tone = torch.LongTensor(tone) + language = torch.LongTensor(language) + + return bert, phone, tone, language + +def infer(text, sdp_ratio, noise_scale, noise_scale_w,length_scale,sid): + bert, phones, tones, lang_ids = get_text(text,"ZH", hps,) + with torch.no_grad(): + x_tst=phones.to(dev).unsqueeze(0) + tones=tones.to(dev).unsqueeze(0) + lang_ids=lang_ids.to(dev).unsqueeze(0) + bert = bert.to(dev).unsqueeze(0) + x_tst_lengths = torch.LongTensor([phones.size(0)]).to(dev) + speakers = torch.LongTensor([hps.data.spk2id[sid]]).to(dev) + 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() + return audio + +def replace_punctuation(text, i=2): + punctuation = "，。？！" + for char in punctuation: + text = text.replace(char, char * i) + return text # Load Generator -hps_mt = utils.get_hparams_from_file("/GPUFS/sysu_hpcedu_123/vits/configs/genshin_xm37.json") +hps = utils.get_hparams_from_file("./configs/config.json") -net_g_mt = SynthesizerTrn( +dev='cuda' +net_g = SynthesizerTrn( len(symbols), - hps_mt.data.filter_length // 2 + 1, - hps_mt.train.segment_size // hps_mt.data.hop_length, - n_speakers=hps_mt.data.n_speakers, - **hps_mt.model).cuda() -_ = net_g_mt.eval() + hps.data.filter_length // 2 + 1, + hps.train.segment_size // hps.data.hop_length, + n_speakers=hps.data.n_speakers, + **hps.model).to(dev) +_ = net_g.eval() -_ = utils.load_checkpoint("/GPUFS/sysu_hpcedu_123/vits/logs/xm37/G_xm37_361200.pth", net_g_mt, None) - -npcList = ['空', '荧', '派蒙', '纳西妲', '阿贝多', '温迪', '枫原万叶', '钟离', '荒泷一斗', '八重神子', '艾尔海森', '提纳里', '迪希雅', '卡维', '宵宫', '莱依拉', '赛诺', '诺艾尔', '托马', '凝光', '莫娜', '北斗', '神里绫华', '雷电将军', '芭芭拉', '鹿野院平藏', '五郎', '迪奥娜', '凯亚', '安柏', '班尼特', '琴', '柯莱', '夜兰', '妮露', '辛焱', '珐露珊', '魈', '香菱', '达达利亚', '砂糖', '早柚', '云堇', '刻晴', '丽莎', '迪卢克', '烟绯', '重云', '珊瑚宫心海', '胡桃', '可莉', '流浪者', '久岐忍', '神里绫人', '甘雨', '戴因斯雷布', '优菈', '菲谢尔', '行秋', '白术', '九条裟罗', '雷泽', '申鹤', '迪娜泽黛', '凯瑟琳', '多莉', '坎蒂丝', '萍姥姥', '罗莎莉亚', '留云借风真君', '绮良良', '瑶瑶', '七七', '奥兹', '米卡', '夏洛蒂', '埃洛伊', '博士', '女士', '大慈树王', '三月七', '娜塔莎', '希露瓦', '虎克', '克拉拉', '丹恒', '希儿', '布洛妮娅', '瓦尔特', '杰帕德', '佩拉', '姬子', '艾丝妲', '白露', '星', '穹', '桑博', '伦纳德', '停云', '罗刹', '卡芙卡', '彦卿', '史瓦罗', '螺丝咕姆', '阿兰', '银狼', '素裳', '丹枢', '黑塔', '景元', '帕姆', '可可利亚', '半夏', '符玄', '公输师傅', '奥列格', '青雀', '大毫', '青镞', '费斯曼', '绿芙蓉', '镜流', '信使', '丽塔', '失落迷迭', '缭乱星棘', '伊甸', '伏特加女孩', '狂热蓝调', '莉莉娅', '萝莎莉娅', '八重樱', '八重霞', '卡莲', '第六夜想曲', '卡萝尔', '姬子', '极地战刃', '布洛妮娅', '次生银翼', '理之律者', '真理之律者', '迷城骇兔', '希儿', '魇夜星渊', '黑希儿', '帕朵菲莉丝', '天元骑英', '幽兰黛尔', '德丽莎', '月下初拥', '朔夜观星', '暮光骑士', '明日香', '李素裳', '格蕾修', '梅比乌斯', '渡鸦', '人之律者', '爱莉希雅', '爱衣', '天穹游侠', '琪亚娜', '空之律者', '终焉之律者', '薪炎之律者', '云墨丹心', '符华', '识之律者', '维尔薇', '始源之律者', '芽衣', '雷之律者', '苏莎娜', '阿波尼亚', '陆景和', '莫弈', '夏彦', '左然', '标贝'] +_ = utils.load_checkpoint("logs/all_in_one/G_521000.pth", net_g, None) @app.route("/",methods=['GET','POST']) def main(): @@ -54,9 +88,10 @@ def main(): try: speaker = request.args.get('speaker') text = request.args.get('text').replace("/n","") + sdp_ratio = float(request.args.get("sdp_ratio", 0.2)) noise = float(request.args.get("noise", 0.5)) noisew = float(request.args.get("noisew", 0.6)) - length = float(request.args.get("length", 1.3)) + length = float(request.args.get("length", 1.2)) if length >= 2: return "Too big length" if len(text) >=200: @@ -69,16 +104,11 @@ def main(): except: return "Invalid Parameter" - stn_tst_mt = get_text(text, hps_mt) with torch.no_grad(): - x_tst_mt = stn_tst_mt.cuda().unsqueeze(0) - x_tst_mt_lengths = torch.LongTensor([stn_tst_mt.size(0)]).cuda() - sid_mt = torch.LongTensor([npcList.index(speaker)]).cuda() - audio_mt = net_g_mt.infer(x_tst_mt, x_tst_mt_lengths, sid=sid_mt, noise_scale=noise, noise_scale_w=noisew, length_scale=length)[0][0,0].data.cpu().float().numpy() - + audio = infer(text, sdp_ratio=sdp_ratio, noise_scale=noise, noise_scale_w=noisew, length_scale=length, sid=speaker) wav = BytesIO() - wavfile.write(wav, hps_mt.data.sampling_rate, audio_mt) + wavfile.write(wav, hps.data.sampling_rate, audio) torch.cuda.empty_cache() if fmt == "mp3": process = ( @@ -89,22 +119,4 @@ def main(): ) out, _ = process.communicate(input=wav.read()) return Response(out, mimetype="audio/mpeg") - return Response(wav.read(), mimetype="audio/wav") - elif request.method == 'POST': - receive = request.get_data(as_text=True) - data = json.loads(receive) - speaker = data["speaker"] - text = data["text"].replace("/n","") - stn_tst_mt = get_text(text, hps_mt) - - with torch.no_grad(): - x_tst_mt = stn_tst_mt.cuda().unsqueeze(0) - x_tst_mt_lengths = torch.LongTensor([stn_tst_mt.size(0)]).cuda() - sid_mt = torch.LongTensor([npcList.index(speaker)]).cuda() - audio_mt = net_g_mt.infer(x_tst_mt, x_tst_mt_lengths, sid=sid_mt, noise_scale=0.667, noise_scale_w=0.8, length_scale=1.15)[0][0,0].data.cpu().float().numpy() - - wav = BytesIO() - wavfile.write(wav, hps_mt.data.sampling_rate, audio_mt) - torch.cuda.empty_cache() - return Response(base64.b64encode(wav.read())) - #return Response(wav.read(), mimetype="audio/wav") + return Response(wav.read(), mimetype="audio/wav") \ No newline at end of file