Update app.py
Browse files
app.py
CHANGED
@@ -1,3 +1,32 @@
|
|
1 |
import gradio as gr
|
|
|
|
|
|
|
2 |
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
import torchaudio
|
3 |
+
from speechbrain.inference.enhancement import WaveformEnhancement
|
4 |
+
import torch
|
5 |
|
6 |
+
# Load the SpeechBrain enhancement model
|
7 |
+
enhance_model = WaveformEnhancement.from_hparams(
|
8 |
+
source="speechbrain/mtl-mimic-voicebank",
|
9 |
+
savedir="pretrained_models/mtl-mimic-voicebank",
|
10 |
+
)
|
11 |
+
|
12 |
+
def enhance_audio(input_audio):
|
13 |
+
# Load the uploaded audio file
|
14 |
+
waveform, sample_rate = torchaudio.load(input_audio)
|
15 |
+
|
16 |
+
# Enhance the audio
|
17 |
+
enhanced_waveform = enhance_model.enhance_batch(waveform)
|
18 |
+
|
19 |
+
# Save the enhanced audio to a file
|
20 |
+
output_path = "enhanced_audio.wav"
|
21 |
+
torchaudio.save(output_path, enhanced_waveform.cpu(), sample_rate)
|
22 |
+
|
23 |
+
return output_path
|
24 |
+
|
25 |
+
# Set up the Gradio interface
|
26 |
+
demo = gr.Interface(
|
27 |
+
fn=enhance_audio,
|
28 |
+
inputs=gr.Audio(type="filepath"), # Upload an audio file
|
29 |
+
outputs=gr.Audio(type="filepath"), # Download the enhanced audio
|
30 |
+
)
|
31 |
+
|
32 |
+
demo.launch()
|