diablofx's picture
Update app.py
d0dbc19
raw
history blame
3.34 kB
import gradio as gr
import numpy as np
import os
import soundfile as sf
def main():
# Gradio Interface
with gr.Blocks() as app:
gr.Markdown(
"""
# <div align="center"> diablofx Audio Interval Cutter (BETA) </div>
Want to [support](https://ko-fi.com/diablofx) me? Or [join AI HUB](https://discord.gg/aihub) for more Help!\n
"""
)
with gr.Row():
with gr.Column():
audio_input = gr.Audio(type='filepath')
get_info_butt = gr.Button(value='Get Audio File Info', variant='primary')
interval_input = gr.Number(value=5, label="Enter Interval (seconds):", min=1, step=1)
with gr.Column():
output_markdown = gr.Markdown(value="", visible=True)
cut_audio_butt = gr.Button(value='Cut and Compile Audio', variant='success', disabled=True)
get_info_butt.click(fn=get_audio_file_info, inputs=[audio_input, interval_input], outputs=[output_markdown, cut_audio_butt])
cut_audio_butt.click(fn=cut_and_compile_audio, inputs=[audio_input, interval_input], outputs=[output_markdown])
app.queue(max_size=1022).launch(share=True)
def get_audio_file_info(audio_file, interval):
# Read the audio data from the file
audio_data, sample_rate = sf.read(audio_file)
# Convert to mono if it's not mono
if len(audio_data.shape) > 1:
audio_data = np.mean(audio_data, axis=1)
# Get the audio file info
audio_info = sf.info(audio_file)
bit_depth = {'PCM_16': 16, 'FLOAT': 32}.get(audio_info.subtype, 0)
# Convert duration to minutes and seconds
minutes, seconds = divmod(audio_info.duration, 60)
# Convert bitrate to kbps
speed_in_kbps = audio_info.samplerate * bit_depth / 1000
# Create a table with the audio file info
info_table = f"""
| Information | Value |
| :---: | :---: |
| File Name | {os.path.basename(audio_file)} |
| Duration | {int(minutes)} minutes - {int(seconds)} seconds |
| Bitrate | {speed_in_kbps} kbp/s |
| Audio Channels | {audio_info.channels} |
| Samples per second | {audio_info.samplerate} Hz |
"""
# Enable the button for cutting audio
return info_table
def cut_and_compile_audio(audio_file, interval):
# Read the audio data from the file
audio_data, sample_rate = sf.read(audio_file)
# Convert to mono if it's not mono
if len(audio_data.shape) > 1:
audio_data = np.mean(audio_data, axis=1)
# Calculate the number of frames for the specified interval
frames_per_interval = int(sample_rate * interval)
# Cut the audio into intervals
audio_clips = [audio_data[i:i + frames_per_interval] for i in range(0, len(audio_data), frames_per_interval)]
# Compile audio clips into a single array
compiled_audio = np.concatenate(audio_clips)
# Save the compiled audio to a new file
compiled_file_path = 'compiled_audio.wav'
sf.write(compiled_file_path, compiled_audio, sample_rate)
# Provide a download link for the compiled audio
download_link = f"[Download Compiled Audio](sandbox:/mnt/data/{compiled_file_path})"
return f"Audio cut and compiled successfully. {download_link}"
# Create the Gradio interface
main()