Spaces:
Running
on
Zero
Running
on
Zero
File size: 2,912 Bytes
02c7bdf bdaf47a 02c7bdf b78b7d0 02c7bdf b78b7d0 a9e592e 132a2a9 f821359 02c7bdf 3192961 b78b7d0 3192961 3956066 b78b7d0 e805751 b78b7d0 e805751 b78b7d0 3956066 e805751 3956066 e805751 3956066 b78b7d0 3956066 95c0d52 d04ae35 3956066 b78b7d0 99710ec b78b7d0 99710ec |
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 68 69 |
import torch
import soundfile as sf
import gradio as gr
from clearvoice import ClearVoice
#myClearVoice = ClearVoice(task='speech_enhancement', model_names=['FRCRN_SE_16K'])
def fn_clearvoice_se(input_wav):
myClearVoice = ClearVoice(task='speech_enhancement', model_names=['FRCRN_SE_16K'])
output_wav_dict = myClearVoice(input_path=input_wav, online_write=False)
if isinstance(output_wav_dict, dict):
key = next(iter(output_wav_dict))
output_wav = output_wav_dict[key]
else:
output_wav = output_wav_dict
sf.write('enhanced.wav', output_wav, 16000)
return 'enhanced.wav'
def fn_clearvoice_ss(input_wav):
myClearVoice = ClearVoice(task='speech_separation', model_names=['MossFormer2_SS_16K'])
output_wav_dict = myClearVoice(input_path=input_wav, online_write=False)
if isinstance(output_wav_dict, dict):
key = next(iter(output_wav_dict))
output_wav = output_wav_dict[key]
else:
output_wav = output_wav_dict
sf.write('separated_1.wav', output_wav, 16000)
return 'separated_1.wav'
demo = gr.Blocks()
se_demo = gr.Interface(
fn=fn_clearvoice_se,
inputs = [
gr.Audio(label="Input Audio", type="filepath"),
],
outputs = [
gr.Audio(label="Output Audio", type="filepath"),
],
title = "ClearVoice: Speech Enhancement",
description = ("Gradio demo for Speech enhancement with ClearVoice. To use it, simply upload your audio, or click one of the examples to load them. Read more at the links below."),
article = ("<p style='text-align: center'><a href='https://arxiv.org/abs/2206.07293' target='_blank'>FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement</a> | <a href='https://github.com/speechbrain/speechbrain' target='_blank'>Github Repo</a></p>"),
examples = [
['mandarin_speech.wav']
],
cache_examples = True,
)
ss_demo = gr.Interface(
fn=fn_clearvoice,
inputs = [
gr.Audio(label="Input Audio", type="filepath"),
],
outputs = [
gr.Audio(label="Output Audio", type="filepath"),
gr.Audio(label="Output Audio", type="filepath"),
],
title = "ClearVoice: Speech Separation",
description = ("Gradio demo for Speech enhancement with ClearVoice. To use it, simply upload your audio, or click one of the examples to load them. Read more at the links below."),
article = ("<p style='text-align: center'><a href='https://arxiv.org/abs/2206.07293' target='_blank'>FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement</a> | <a href='https://github.com/speechbrain/speechbrain' target='_blank'>Github Repo</a></p>"),
examples = [
['mandarin_speech.wav']
],
cache_examples = True,
)
with demo:
gr.TabbedInterface([se_demo, ss_demo], ["Speech Enhancement", "Speech Separation"])
demo.launch() |