app.py

import os
import glob
import json
import argparse
import traceback
import logging
import gradio as gr
import numpy as np
import librosa
import torch
import asyncio
import edge_tts
from datetime import datetime
from fairseq import checkpoint_utils
from infer_pack.models import SynthesizerTrnMs256NSFsid, SynthesizerTrnMs256NSFsid_nono
from vc_infer_pipeline import VC
from config import Config
config = Config()
logging.getLogger("numba").setLevel(logging.WARNING)
limitation = os.getenv("SYSTEM") == "spaces"  # limit audio length in huggingface spaces

def create_vc_fn(tgt_sr, net_g, vc, if_f0, file_index):
    def vc_fn(
        input_audio,
        f0_up_key,
        f0_method,
        index_rate,
        tts_mode,
        tts_text,
        tts_voice
    ):
        try:
            if tts_mode:
                if len(tts_text) > 100 and limitation:
                    return "Text is too long", None
                if tts_text is None or tts_voice is None:
                    return "You need to enter text and select a voice", None
                asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save("tts.mp3"))
                audio, sr = librosa.load("tts.mp3", sr=16000, mono=True)
            else:
                if input_audio is None:
                    return "You need to upload an audio", None
                sampling_rate, audio = input_audio
                duration = audio.shape[0] / sampling_rate
                if duration > 20 and limitation:
                    return "Please upload an audio file that is less than 20 seconds. If you need to generate a longer audio file, please use Colab.", None
                audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
                if len(audio.shape) > 1:
                    audio = librosa.to_mono(audio.transpose(1, 0))
                if sampling_rate != 16000:
                    audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
            times = [0, 0, 0]
            f0_up_key = int(f0_up_key)
            audio_opt = vc.pipeline(
                hubert_model,
                net_g,
                0,
                audio,
                times,
                f0_up_key,
                f0_method,
                file_index,
                index_rate,
                if_f0,
                f0_file=None,
            )
            print(
                f"[{datetime.now().strftime('%Y-%m-%d %H:%M')}]: npy: {times[0]}, f0: {times[1]}s, infer: {times[2]}s"
            )
            return "Success", (tgt_sr, audio_opt)
        except:
            info = traceback.format_exc()
            print(info)
            return info, (None, None)
    return vc_fn

def load_hubert():
    global hubert_model
    models, _, _ = checkpoint_utils.load_model_ensemble_and_task(
        ["hubert_base.pt"],
        suffix="",
    )
    hubert_model = models[0]
    hubert_model = hubert_model.to(config.device)
    if config.is_half:
        hubert_model = hubert_model.half()
    else:
        hubert_model = hubert_model.float()
    hubert_model.eval()

def change_to_tts_mode(tts_mode):
    if tts_mode:
        return gr.Audio.update(visible=False), gr.Textbox.update(visible=True), gr.Dropdown.update(visible=True)
    else:
        return gr.Audio.update(visible=True), gr.Textbox.update(visible=False), gr.Dropdown.update(visible=False)

if __name__ == '__main__':
    load_hubert()
    models = []
    tts_voice_list = asyncio.get_event_loop().run_until_complete(edge_tts.list_voices())
    voices = [f"{v['ShortName']}-{v['Gender']}" for v in tts_voice_list]
    if limitation:
        with open("weights/model_info.json", "r", encoding="utf-8") as f:
            models_info = json.load(f)
        for name, info in models_info.items():
            if not info['enable']:
                continue
            title = info['title']
            author = info.get("author", None)
            cover = f"weights/{name}/{info['cover']}"
            index = f"weights/{name}/{info['feature_retrieval_library']}"
            cpt = torch.load(f"weights/{name}/{name}.pth", map_location="cpu")
            tgt_sr = cpt["config"][-1]
            cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0]  # n_spk
            if_f0 = cpt.get("f0", 1)
            if if_f0 == 1:
                net_g = SynthesizerTrnMs256NSFsid(*cpt["config"], is_half=config.is_half)
            else:
                net_g = SynthesizerTrnMs256NSFsid_nono(*cpt["config"])
            del net_g.enc_q
            print(net_g.load_state_dict(cpt["weight"], strict=False))  # 不加这一行清不干净, 真奇葩
            net_g.eval().to(config.device)
            if config.is_half:
                net_g = net_g.half()
            else:
                net_g = net_g.float()
            vc = VC(tgt_sr, config)
            models.append((name, title, author, cover, create_vc_fn(tgt_sr, net_g, vc, if_f0, index)))
    else:
        folder_path = "weights"
        for name in os.listdir(folder_path):
            print("check folder: " + name)
            if name.startswith("."): break
            cover_path = glob.glob(f"{folder_path}/{name}/*.png") + glob.glob(f"{folder_path}/{name}/*.jpg")
            index_path = glob.glob(f"{folder_path}/{name}/*.index")
            checkpoint_path = glob.glob(f"{folder_path}/{name}/*.pth")
            title = name
            author = ""
            if cover_path:
                cover = cover_path[0]
            else:
                cover = ""
            index = index_path[0]
            cpt = torch.load(checkpoint_path[0], map_location="cpu")
            tgt_sr = cpt["config"][-1]
            cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0]  # n_spk
            if_f0 = cpt.get("f0", 1)
            if if_f0 == 1:
                net_g = SynthesizerTrnMs256NSFsid(*cpt["config"], is_half=config.is_half)
            else:
                net_g = SynthesizerTrnMs256NSFsid_nono(*cpt["config"])
            del net_g.enc_q
            print(net_g.load_state_dict(cpt["weight"], strict=False))  # 不加这一行清不干净, 真奇葩
            net_g.eval().to(config.device)
            if config.is_half:
                net_g = net_g.half()
            else:
                net_g = net_g.float()
            vc = VC(tgt_sr, config)
            models.append((name, title, author, cover, create_vc_fn(tgt_sr, net_g, vc, if_f0, index)))
    with gr.Blocks() as app:
        gr.Markdown(
            "# <center> RVC Models (Latest Update)\n"
            "## <center> The input audio should be clean and pure voice without background music.\n"
            "### <center> [Recommended to use google colab for more features](https://colab.research.google.com/drive/110kiMZTdP6Ri1lY9-NbQf17GVPPhHyeT?usp=sharing) \n"
            "#### <center> Please regenerate your model to latest RVC to fully applied this new rvc.\n"
            "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/110kiMZTdP6Ri1lY9-NbQf17GVPPhHyeT?usp=sharing)\n\n"
            "[![Original Repo](https://badgen.net/badge/icon/github?icon=github&label=Original%20Repo)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)"
        )
        with gr.Tabs():
            for (name, title, author, cover, vc_fn) in models:
                with gr.TabItem(name):
                    with gr.Row():
                        gr.Markdown(
                            '<div align="center">'
                            f'<div>{title}</div>\n'+
                            (f'<div>Model author: {author}</div>' if author else "")+
                            (f'<img style="width:auto;height:300px;" src="file/{cover}">' if cover else "")+
                            '</div>'
                        )
                    with gr.Row():
                        with gr.Column():
                            vc_input = gr.Audio(label="Input audio"+' (less than 20 seconds)' if limitation else '')
                            vc_transpose = gr.Number(label="Transpose", value=0)
                            vc_f0method = gr.Radio(
                                label="Pitch extraction algorithm, PM is fast but Harvest is better for low frequencies",
                                choices=["pm", "harvest"],
                                value="pm",
                                interactive=True,
                            )
                            vc_index_ratio = gr.Slider(
                                minimum=0,
                                maximum=1,
                                label="Retrieval feature ratio",
                                value=0.6,
                                interactive=True,
                            )
                            tts_mode = gr.Checkbox(label="tts (use edge-tts as input)", value=False)
                            tts_text = gr.Textbox(visible=False,label="TTS text (100 words limitation)" if limitation else "TTS text")
                            tts_voice = gr.Dropdown(label="Edge-tts speaker", choices=voices, visible=False, allow_custom_value=False, value="en-US-AnaNeural-Female")
                            vc_submit = gr.Button("Generate", variant="primary")
                        with gr.Column():
                            vc_output1 = gr.Textbox(label="Output Message")
                            vc_output2 = gr.Audio(label="Output Audio")
                vc_submit.click(vc_fn, [vc_input, vc_transpose, vc_f0method, vc_index_ratio, tts_mode, tts_text, tts_voice], [vc_output1, vc_output2])
                tts_mode.change(change_to_tts_mode, [tts_mode], [vc_input, tts_text, tts_voice])
        gr.Markdown('# <center>Changelog 2023.05.15')
        gr.Markdown('- Added support for direct upload to gradio')
        gr.Markdown('- Added ayato-jp, eula-jp, and keqing-jp')
        gr.Markdown('- Minor fix and adjustment')
        app.queue(concurrency_count=1, max_size=20, api_open=config.api).launch(share=config.colab)