Amarsaish commited on
Commit
8cf577a
·
verified ·
1 Parent(s): e1695d5

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +76 -7
app.py CHANGED
@@ -1,12 +1,80 @@
1
- import os
2
- import streamlit as st
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  from pydub import AudioSegment
4
  from groq import Groq
5
-
6
- # Set ffmpeg path
7
- ffmpeg_path = r"ffmpeg.exe"
8
- os.environ["PATH"] += os.pathsep + os.path.dirname(ffmpeg_path)
9
- AudioSegment.converter = ffmpeg_path
10
 
11
  # Groq API configuration
12
  groq_api_key = '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' # Replace with your actual API key
@@ -69,3 +137,4 @@ if uploaded_file:
69
  st.text_area("Transcribed Text", transcription, height=200)
70
  except Exception as e:
71
  st.error(f"Error during transcription: {e}")
 
 
1
+ # import os
2
+ # import streamlit as st
3
+ # from pydub import AudioSegment
4
+ # from groq import Groq
5
+
6
+ # # Set ffmpeg path
7
+ # ffmpeg_path = r"ffmpeg.exe"
8
+ # os.environ["PATH"] += os.pathsep + os.path.dirname(ffmpeg_path)
9
+ # AudioSegment.converter = ffmpeg_path
10
+
11
+ # # Groq API configuration
12
+ # groq_api_key = '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' # Replace with your actual API key
13
+ # client = Groq(api_key=groq_api_key)
14
+ # model = 'whisper-large-v3'
15
+
16
+ # # Function to ensure the file is in a suitable format
17
+ # def ensure_suitable_format(file_path):
18
+ # allowed_formats = ["flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "opus", "wav", "webm"]
19
+ # file_extension = file_path.split('.')[-1].lower()
20
+ # if file_extension not in allowed_formats:
21
+ # new_file_path = f"{os.path.splitext(file_path)[0]}.wav"
22
+ # os.rename(file_path, new_file_path)
23
+ # return new_file_path
24
+ # return file_path
25
+
26
+ # # Function to convert audio to WAV
27
+ # def convert_audio_to_wav(input_path, output_path):
28
+ # audio = AudioSegment.from_file(input_path)
29
+ # audio.export(output_path, format="wav")
30
+ # return output_path
31
+
32
+ # # Function to transcribe audio using Groq
33
+ # def audio_to_text(filepath):
34
+ # with open(filepath, "rb") as file:
35
+ # translation = client.audio.translations.create(
36
+ # file=(filepath, file.read()),
37
+ # model=model,
38
+ # )
39
+ # return translation.text
40
+
41
+ # # Streamlit App UI
42
+ # st.title("Audio-to-Text Transcription")
43
+ # st.write("Upload an audio file to get the transcribed text.")
44
+
45
+ # # File upload
46
+ # uploaded_file = st.file_uploader("Upload your audio file", type=["flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "opus", "wav", "webm"])
47
+
48
+ # if uploaded_file:
49
+ # # Save the uploaded file locally
50
+ # file_path = os.path.join("uploaded_audio", uploaded_file.name)
51
+ # os.makedirs("uploaded_audio", exist_ok=True)
52
+ # with open(file_path, "wb") as f:
53
+ # f.write(uploaded_file.getbuffer())
54
+
55
+ # st.write(f"File uploaded: {uploaded_file.name}")
56
+
57
+ # # Ensure file format is suitable
58
+ # suitable_audio_path = ensure_suitable_format(file_path)
59
+
60
+ # # Convert audio to WAV
61
+ # wav_path = f"{os.path.splitext(suitable_audio_path)[0]}.wav"
62
+ # converted_audio = convert_audio_to_wav(suitable_audio_path, wav_path)
63
+
64
+ # # Transcribe audio
65
+ # st.write("Processing transcription...")
66
+ # try:
67
+ # transcription = audio_to_text(converted_audio)
68
+ # st.success("Transcription complete!")
69
+ # st.text_area("Transcribed Text", transcription, height=200)
70
+ # except Exception as e:
71
+ # st.error(f"Error during transcription: {e}")
72
+
73
+
74
  from pydub import AudioSegment
75
  from groq import Groq
76
+ import os
77
+ import streamlit as st
 
 
 
78
 
79
  # Groq API configuration
80
  groq_api_key = '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' # Replace with your actual API key
 
137
  st.text_area("Transcribed Text", transcription, height=200)
138
  except Exception as e:
139
  st.error(f"Error during transcription: {e}")
140
+