colab_flask.py

import os
import time
import cv2
from flask import Flask, request, jsonify, send_file
from flask_ngrok2 import run_with_ngrok
from pyngrok import ngrok
from subprocess import call
from tqdm import tqdm
import numpy as np
from ffmpy import FFmpeg

# !pip install flask flask-ngrok2 pyngrok

app = Flask(__name__)
auth_token = '2XQTtjJqVg4B4ryQVgauTPIGeiK_3JUWVJcMhXwpPxz2sc9KB'
root_dir = '/content/Wav2Lip-GFPGAN'
jobs_dir = os.path.join('/content', 'jobs')


@app.route('/', methods=['GET'])
def index():
    return jsonify({'hello': 'world!'})


# New route to serve files from /content/input
@app.route('/job/<job_id>/<filename>', methods=['GET'])
def download_file(job_id, filename):
    try:
        file_path = os.path.join(jobs_dir, job_id, filename)
        print(file_path)
        return send_file(file_path, as_attachment=True)
    except Exception as e:
        return jsonify({'error': str(e)}), 500


@app.route('/wav2lip', methods=['POST'])
def wav2lip():
    try:
        # Get uploaded files
        video_file = request.files['video']
        audio_file = request.files['audio']
        job_id = str(int(time.time()))
        job_path = os.path.join(jobs_dir, job_id)
        os.makedirs(job_path, exist_ok=True)

        # Save the files to a temporary directory
        video_path = os.path.join(job_path, video_file.filename)
        audio_path = os.path.join(job_path, audio_file.filename)
        video_file.save(video_path)
        audio_file.save(audio_path)

        wav2lip_mp4 = os.path.join(job_path, 'wav2lip.mp4')
        call_wav2lip(video_path, audio_path, wav2lip_mp4)

        output_filename = 'output.mp4'
        output_mp4 = os.path.join(job_path, output_filename)
        call_gfpgan(wav2lip_mp4, audio_path, output_mp4)

        ngrok_url = ngrok.get_tunnels()[0].public_url
        return jsonify({'url': f'{ngrok_url}/job/{job_id}/{output_filename}'})

    except Exception as e:
        return jsonify({'error': str(e)}), 500


def call_wav2lip(video_path, audio_path, output_path):
    checkpoint_path = os.path.join(root_dir, 'wav2lip/checkpoints/wav2lip.pth')
    assert os.path.isfile(video_path), f'Video path {video_path} not exist.'
    assert os.path.isfile(audio_path), f'Audio path {audio_path} not exist.'
    assert os.path.isfile(checkpoint_path), f'Checkpoint file {checkpoint_path} not exist.'

    # python inference.py \
    # --checkpoint_path checkpoints/wav2lip.pth \
    # --face {inputVideoPath} \
    # --audio {inputAudioPath} \
    # --outfile {lipSyncedOutputPath}
    cmd = [
        "python",
        "wav2lip/inference.py",
        "--checkpoint_path", checkpoint_path,
        # "--segmentation_path", "checkpoints/face_segmentation.pth",
        "--face", video_path,
        "--audio", audio_path,
        "--outfile", output_path,
    ]

    call(cmd)
    return output_path


def _get_frames(video_path):
    folder_path = os.path.dirname(video_path)
    origin_frames_folder = os.path.join(folder_path, 'frames')
    os.makedirs(origin_frames_folder, exist_ok=True)

    # get frames pics
    vidcap = cv2.VideoCapture(video_path)
    numberOfFrames = int(vidcap.get(cv2.CAP_PROP_FRAME_COUNT))
    fps = vidcap.get(cv2.CAP_PROP_FPS)
    print("FPS: ", fps, "Frames: ", numberOfFrames)

    for frameNumber in tqdm(range(numberOfFrames)):
        _, image = vidcap.read()
        cv2.imwrite(os.path.join(origin_frames_folder, str(frameNumber).zfill(4) + '.jpg'), image)
    return origin_frames_folder


def call_gfpgan(wav2lip_mp4, audio_path, output_mp4):
    assert os.path.isfile(wav2lip_mp4), f'Video path {wav2lip_mp4} not exist.'
    origin_frames_folder = _get_frames(wav2lip_mp4)
    folder_path = os.path.dirname(wav2lip_mp4)

    # python inference_gfpgan.py
    # -i "$unProcessedFramesFolderPath"
    # -o "$outputPath"
    # -v 1.3
    # -s 2
    # --only_center_face
    # --bg_upsampler None
    cmd = [
        "python",
        "gfpgan/inference_gfpgan.py",
        "-i", origin_frames_folder,
        "-o", folder_path,
        "-v", 1.3,
        "-s", 2,
        "--only_center_face",
        "--bg_upsampler", None
    ]
    call(cmd)

    restoredFramesPath = os.path.join(folder_path, 'restored_imgs')
    os.makedirs(restoredFramesPath, exist_ok=True)
    folder_path = folder_path
    dir_list = os.listdir(restoredFramesPath)
    dir_list.sort()
    batch = 0
    batchSize = 300
    for i in tqdm(range(0, len(dir_list), batchSize)):
        img_array = []
        start, end = i, i + batchSize
        print("processing ", start, end)
        for filename in tqdm(dir_list[start:end]):
            filename = restoredFramesPath + filename;
            img = cv2.imread(filename)
            if img is None:
                continue
            height, width, layers = img.shape
            size = (width, height)
            img_array.append(img)

        fourcc = cv2.VideoWriter_fourcc(*'X264')
        filename = 'batch_' + str(batch).zfill(4) + '.mp4'
        out = cv2.VideoWriter(os.path.join(folder_path, filename),
                              fourcc, 30, size)
        batch = batch + 1

        for i in range(len(img_array)):
            out.write(img_array[i])
        out.release()

    concatTextFilePath = os.path.join(folder_path, "concat.txt")

    concatTextFile = open(concatTextFilePath, "w")
    for ips in range(batch):
        concatTextFile.write("file batch_" + str(ips).zfill(4) + ".mp4\n")
    concatTextFile.close()

    concatedVideoOutputPath = os.path.join(folder_path, "concated_output.mp4")
    ff = FFmpeg(
        inputs={concatTextFilePath: None},
        outputs={concatedVideoOutputPath: '-y -c copy'}
    )
    # !ffmpeg -y -f concat -i {concatTextFilePath} -c copy {concatedVideoOutputPath}
    ff.run()

    ff = FFmpeg(
        inputs={concatedVideoOutputPath: None, audio_path: None},
        outputs={output_mp4: '-y -map 0 -map 1:a -c:v libx264 -c:a aac -shortest'}
    )
    # !ffmpeg -y -i {concatedVideoOutputPath} -i {inputAudioPath} -map 0 -map 1:a -c:v copy -shortest {finalProcessedOuputVideo}
    ff.run()

    # from google.colab import files
    # files.download(finalProcessedOuputVideo)


if __name__ == '__main__':
    ngrok_url = run_with_ngrok(app, auth_token=auth_token).public_url  # Start ngrok when the app is run
    print(f" * Running on {ngrok_url}")

    app.run()


def test():
    # request
    import requests
    ngrok_url = f"http://74c0-34-87-172-60.ngrok-free.app"
    url = f"{ngrok_url}/wav2lip"
    print(url)
    video_path = '/Users/taoluo/Downloads/oIy5B4-vHVw.4.6588496370531551262.0.jpg'
    audio_path = '/Users/taoluo/Downloads/test_audio.mp3'
    files = {'video': ('video.jpg', open(video_path, 'rb')), 'audio': ('audio.mp3', open(audio_path, 'rb'))}
    headers = {'ngrok-skip-browser-warning': 'true'}
    response = requests.post(url, files=files, headers=headers)
    # Print the response
    print(response.json())
    data = response.json()
    print(ngrok_url + data['url'])