import gradio as gr
import numpy as np
import os
import soundfile as sf
from pydub import AudioSegment
from io import BytesIO
def main():
# Gradio Interface
with gr.Blocks() as app:
gr.Markdown(
"""
#
diablofx Audio Interval Cutter (BETA)
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')
create_info_butt = gr.Button(value='Get Audio File Info', variant='primary')
split_audio_butt = gr.Button(value='Split Audio into 5s Clips', variant='success')
with gr.Column():
output_markdown = gr.Markdown(value="", visible=True)
create_info_butt.click(fn=get_audio_file_info, inputs=[audio_input], outputs=[output_markdown])
split_audio_butt.click(fn=split_audio_into_clips, inputs=[audio_input])
app.queue(max_size=1022).launch(share=True)
def get_audio_file_info(audio_file):
# 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 |
"""
# Return the info table
return info_table
def split_audio_into_clips(audio_file):
# 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)
# Convert to Pydub AudioSegment
audio_segment = AudioSegment(audio_data.tobytes(), frame_rate=sample_rate, sample_width=audio_data.dtype.itemsize, channels=1)
# Split the audio into 5s clips
clips = [audio_segment[i:i + 5000] for i in range(0, len(audio_segment), 5000)]
# Save each clip and create download buttons
download_buttons = []
for i, clip in enumerate(clips):
clip_data = clip.raw_data
clip_io = BytesIO(clip_data)
clip_name = f"clip_{i+1}.wav"
sf.write(clip_name, clip_io.read(), sample_rate)
download_buttons.append((f"Download Clip {i+1}", clip_name))
return download_buttons
# Create the Gradio interface
main()