Update app.py
Browse files
app.py
CHANGED
@@ -15,21 +15,59 @@ from isNumber import is_number
|
|
15 |
from processDoubles import process_doubles
|
16 |
from replaceWords import replace_words
|
17 |
|
18 |
-
|
|
|
19 |
|
20 |
|
21 |
-
def
|
22 |
# # Process the audio file
|
23 |
-
transcript =
|
24 |
text_value = transcript['text']
|
25 |
processd_doubles=process_doubles(text_value)
|
26 |
replaced_words = replace_words(processd_doubles)
|
27 |
converted_text=text_to_int(replaced_words)
|
28 |
return converted_text
|
29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
demo=gr.Interface(
|
31 |
-
|
|
|
32 |
inputs=[
|
|
|
|
|
33 |
gr.Audio(sources=["microphone","upload"], type="filepath"),
|
34 |
],
|
35 |
outputs=[
|
@@ -38,3 +76,4 @@ demo=gr.Interface(
|
|
38 |
title="Automatic Speech Recognition",
|
39 |
description = "Demo for Automatic Speech Recognition. Use microphone to record speech. Please press Record button. Initially it will take some time to load the model. The recognized text will appear in the output textbox",
|
40 |
).launch()
|
|
|
|
15 |
from processDoubles import process_doubles
|
16 |
from replaceWords import replace_words
|
17 |
|
18 |
+
transcriber_hindi_new = pipeline(task="automatic-speech-recognition", model="cdactvm/w2v-bert-2.0-hindi_v1")
|
19 |
+
transcriber_hindi_old = pipeline(task="automatic-speech-recognition", model="cdactvm/w2v-bert-2.0-hindi_old")
|
20 |
|
21 |
|
22 |
+
def transcribe_hindi_new(audio):
|
23 |
# # Process the audio file
|
24 |
+
transcript = transcriber_hindi_new(audio)
|
25 |
text_value = transcript['text']
|
26 |
processd_doubles=process_doubles(text_value)
|
27 |
replaced_words = replace_words(processd_doubles)
|
28 |
converted_text=text_to_int(replaced_words)
|
29 |
return converted_text
|
30 |
|
31 |
+
def transcribe_hindi_old(audio):
|
32 |
+
# # Process the audio file
|
33 |
+
transcript = transcriber_hindi_old(audio)
|
34 |
+
text_value = transcript['text']
|
35 |
+
processd_doubles=process_doubles(text_value)
|
36 |
+
replaced_words = replace_words(processd_doubles)
|
37 |
+
converted_text=text_to_int(replaced_words)
|
38 |
+
return converted_text
|
39 |
+
|
40 |
+
def sel_lng(lng, mic=None, file=None):
|
41 |
+
if mic is not None:
|
42 |
+
audio = mic
|
43 |
+
elif file is not None:
|
44 |
+
audio = file
|
45 |
+
else:
|
46 |
+
return "You must either provide a mic recording or a file"
|
47 |
+
|
48 |
+
if lng == "hindi_old":
|
49 |
+
return transcribe_hindi_old(audio)
|
50 |
+
elif lng == "hindi_new":
|
51 |
+
return transcribe_hindi_new(audio)
|
52 |
+
|
53 |
+
# demo=gr.Interface(
|
54 |
+
# transcribe,
|
55 |
+
# inputs=[
|
56 |
+
# gr.Audio(sources=["microphone","upload"], type="filepath"),
|
57 |
+
# ],
|
58 |
+
# outputs=[
|
59 |
+
# "textbox"
|
60 |
+
# ],
|
61 |
+
# title="Automatic Speech Recognition",
|
62 |
+
# description = "Demo for Automatic Speech Recognition. Use microphone to record speech. Please press Record button. Initially it will take some time to load the model. The recognized text will appear in the output textbox",
|
63 |
+
# ).launch()
|
64 |
+
|
65 |
demo=gr.Interface(
|
66 |
+
fn=sel_lng,
|
67 |
+
|
68 |
inputs=[
|
69 |
+
gr.Dropdown([
|
70 |
+
"hindi_new","hindi_old"],label="Select Language"),
|
71 |
gr.Audio(sources=["microphone","upload"], type="filepath"),
|
72 |
],
|
73 |
outputs=[
|
|
|
76 |
title="Automatic Speech Recognition",
|
77 |
description = "Demo for Automatic Speech Recognition. Use microphone to record speech. Please press Record button. Initially it will take some time to load the model. The recognized text will appear in the output textbox",
|
78 |
).launch()
|
79 |
+
|