Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -2,14 +2,13 @@ from transformers import pipeline
|
|
2 |
import os
|
3 |
import gradio as gr
|
4 |
import torch
|
5 |
-
from IPython.display import Audio as IPythonAudio
|
6 |
-
from gtts import gTTS
|
7 |
-
import IPython.display as ipd
|
8 |
|
9 |
#Audio to text
|
10 |
-
asr = pipeline(task="automatic-speech-recognition",
|
11 |
-
model="distil-whisper/distil-small.en")
|
12 |
-
#asr=pipeline("automatic-speech-recognition", model="openai/whisper-large-v3")
|
13 |
#Text to text
|
14 |
translator = pipeline(task="translation",
|
15 |
model="facebook/nllb-200-distilled-600M",
|
@@ -23,7 +22,8 @@ def transcribe_speech(filepath):
|
|
23 |
if filepath is None:
|
24 |
gr.Warning("No audio found, please retry.")
|
25 |
return ""
|
26 |
-
|
|
|
27 |
src_lang="eng_Latn",
|
28 |
tgt_lang="hin_Deva")
|
29 |
narrated_text=pipe(output[0]['translation_text'])
|
@@ -35,16 +35,18 @@ def transcribe_speech(filepath):
|
|
35 |
|
36 |
mic_transcribe = gr.Interface(
|
37 |
fn=transcribe_speech,
|
38 |
-
inputs=gr.Audio(sources="microphone",
|
39 |
-
|
|
|
40 |
outputs="audio",
|
41 |
#outputs=gr.Audio(label="Translated Message"),
|
42 |
allow_flagging="never")
|
43 |
|
44 |
file_transcribe = gr.Interface(
|
45 |
fn=transcribe_speech,
|
46 |
-
inputs=gr.Audio(sources="upload",
|
47 |
-
|
|
|
48 |
outputs="audio",
|
49 |
#outputs=gr.Audio(label="Translated Message"),
|
50 |
allow_flagging="never"
|
|
|
2 |
import os
|
3 |
import gradio as gr
|
4 |
import torch
|
5 |
+
#from IPython.display import Audio as IPythonAudio
|
6 |
+
#from gtts import gTTS
|
7 |
+
#import IPython.display as ipd
|
8 |
|
9 |
#Audio to text
|
10 |
+
#asr = pipeline(task="automatic-speech-recognition",
|
11 |
+
# model="distil-whisper/distil-small.en")
|
|
|
12 |
#Text to text
|
13 |
translator = pipeline(task="translation",
|
14 |
model="facebook/nllb-200-distilled-600M",
|
|
|
22 |
if filepath is None:
|
23 |
gr.Warning("No audio found, please retry.")
|
24 |
return ""
|
25 |
+
#asr(filepath)["text"]
|
26 |
+
output = translator(filepath,
|
27 |
src_lang="eng_Latn",
|
28 |
tgt_lang="hin_Deva")
|
29 |
narrated_text=pipe(output[0]['translation_text'])
|
|
|
35 |
|
36 |
mic_transcribe = gr.Interface(
|
37 |
fn=transcribe_speech,
|
38 |
+
#inputs=gr.Audio(sources="microphone",
|
39 |
+
# type="filepath"),
|
40 |
+
inputs=gr.Textbox(label="text",lines=3),
|
41 |
outputs="audio",
|
42 |
#outputs=gr.Audio(label="Translated Message"),
|
43 |
allow_flagging="never")
|
44 |
|
45 |
file_transcribe = gr.Interface(
|
46 |
fn=transcribe_speech,
|
47 |
+
#inputs=gr.Audio(sources="upload",
|
48 |
+
# type="filepath"),
|
49 |
+
inputs=gr.Textbox(label="text",lines=3),
|
50 |
outputs="audio",
|
51 |
#outputs=gr.Audio(label="Translated Message"),
|
52 |
allow_flagging="never"
|