Spaces:
Running
Running
File size: 2,343 Bytes
80ebd45 4e7548f 3428530 1da425c 3428530 1da425c 80ebd45 1da425c 80ebd45 3428530 1da425c 3428530 0a44c9b 1da425c 4e7548f 1da425c 9f9e12c 1da425c 4e7548f 3428530 f700d4a 1b4bfa7 bb24359 1b4bfa7 1da425c 4e7548f 1b4bfa7 4e7548f 1da425c 1b4bfa7 37d0f49 1b4bfa7 4e7548f 1b4bfa7 4e7548f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
import gradio as gr
import spaces
import subprocess
import os
from PIL import Image
import ffmpeg
from pydub import AudioSegment
import numpy as np
import soundfile as sf
def save_audio_mp3(audio_tuple, filename):
sampling_rate, audio_data = audio_tuple
audio_bytes = np.array(audio_data, dtype=np.int16).tobytes()
audio_segment = AudioSegment(audio_bytes, sample_width=2, frame_rate=sampling_rate, channels=1)
audio_segment.export(filename, format="mp3")
return f"Audio saved successfully as {filename}"
def audio_video():
input_video = ffmpeg.input('results/result_voice.mp4')
input_audio = ffmpeg.input('sample_data/uploaded_audio.mp3')
os.system(f"rm -rf results/final_output.mp4")
ffmpeg.concat(input_video, input_audio, v=1, a=1).output('results/final_output.mp4').run()
return "results/final_output.mp4"
@spaces.GPU
def run_infrence(input_image,input_audio):
pil_image = Image.fromarray(input_image.astype(np.uint8))
save_dir = "sample_data"
if not os.path.exists(save_dir):
os.makedirs(save_dir)
# Save input image
filename = os.path.join(save_dir, "uploaded_image.png")
pil_image.save(filename)
#Save input audio
save_audio_mp3(input_audio, "sample_data/uploaded_audio.mp3")
command = f'python3 inference.py --checkpoint_path checkpoints/wav2lip_gan.pth --face sample_data/uploaded_image.png --audio sample_data/uploaded_audio.mp3'
process = subprocess.Popen(command, stdout=subprocess.PIPE, shell=True)
output, error = process.communicate()
return audio_video()
def run():
with gr.Blocks(css=".gradio-container {background-color: lightgray} #radio_div {background-color: #FFD8B4; font-size: 40px;}") as demo:
gr.Markdown("<h1 style='text-align: center;'>"+ "Wav2lip Lip Sync Using Image and Audio" + "</h1><br/><br/>")
with gr.Group():
with gr.Row():
input_image = gr.Image(label="Input Image")
input_audio = gr.Audio(label="Input Audio")
video_out = gr.Video(show_label=True,label="Output")
with gr.Row():
btn = gr.Button("Generate")
btn.click(run_infrence,inputs=[input_image,input_audio], outputs=[video_out])
demo.queue()
demo.launch(server_name="0.0.0.0", server_port=7860)
if __name__ == "__main__":
run() |