Spaces:
Runtime error
Runtime error
Commit
•
62d3d54
1
Parent(s):
d71ae8d
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import pipeline
|
3 |
+
import torch
|
4 |
+
import librosa
|
5 |
+
import soundfile
|
6 |
+
|
7 |
+
SAMPLE_RATE = 16000
|
8 |
+
|
9 |
+
pipe = pipeline(model="birgermoell/whisper-small-sv-bm") # change to "your-username/the-name-you-picked"
|
10 |
+
|
11 |
+
|
12 |
+
def process_audio_file(file):
|
13 |
+
data, sr = librosa.load(file)
|
14 |
+
|
15 |
+
if sr != SAMPLE_RATE:
|
16 |
+
data = librosa.resample(data, sr, SAMPLE_RATE)
|
17 |
+
|
18 |
+
# monochannel
|
19 |
+
data = librosa.to_mono(data)
|
20 |
+
return data
|
21 |
+
|
22 |
+
|
23 |
+
def transcribe(Microphone, File_Upload):
|
24 |
+
warn_output = ""
|
25 |
+
if (Microphone is not None) and (File_Upload is not None):
|
26 |
+
warn_output = "WARNING: You've uploaded an audio file and used the microphone. " \
|
27 |
+
"The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
|
28 |
+
file = Microphone
|
29 |
+
|
30 |
+
elif (Microphone is None) and (File_Upload is None):
|
31 |
+
return "ERROR: You have to either use the microphone or upload an audio file"
|
32 |
+
|
33 |
+
elif Microphone is not None:
|
34 |
+
file = Microphone
|
35 |
+
else:
|
36 |
+
file = File_Upload
|
37 |
+
|
38 |
+
audio_data = process_audio_file(file)
|
39 |
+
|
40 |
+
text = pipe(audio_data)["text"]
|
41 |
+
|
42 |
+
return warn_output + text
|
43 |
+
|
44 |
+
|
45 |
+
iface = gr.Interface(
|
46 |
+
fn=transcribe,
|
47 |
+
inputs=[
|
48 |
+
gr.inputs.Audio(source="microphone", type='filepath', optional=True),
|
49 |
+
gr.inputs.Audio(source="upload", type='filepath', optional=True),
|
50 |
+
],
|
51 |
+
outputs="text",
|
52 |
+
layout="horizontal",
|
53 |
+
theme="huggingface",
|
54 |
+
title="Whisper Small SV",
|
55 |
+
description="Demo for Swedish speech recognition using the [Whisper Small SV BM checkpoint](https://huggingface.co/birgermoell/whisper-small-sv-bm).",
|
56 |
+
allow_flagging='never',
|
57 |
+
)
|
58 |
+
|
59 |
+
iface.launch(enable_queue=True)
|