Spaces:
Runtime error
Runtime error
Commit
•
53978c9
1
Parent(s):
91e6e4f
Create Audio_into_chunks.py
Browse files- Audio_into_chunks.py +56 -0
Audio_into_chunks.py
ADDED
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
from pydub import AudioSegment
|
3 |
+
import whisper
|
4 |
+
from deep_translator import GoogleTranslator
|
5 |
+
|
6 |
+
|
7 |
+
# @title Audio into chunks
|
8 |
+
def audio_into_chunks_transcribe_translate(audio_file,lang):
|
9 |
+
chunk_length_seconds=11
|
10 |
+
output_format="wav"
|
11 |
+
# Check if file exists
|
12 |
+
if not os.path.exists(audio_file):
|
13 |
+
raise ValueError(f"FLAC file not found: {audio_file}")
|
14 |
+
Transcribe_Text=[]
|
15 |
+
# Load the FLAC audio
|
16 |
+
audio_segment = AudioSegment.from_file(audio_file, format="flac")
|
17 |
+
#load Model For Transcribe
|
18 |
+
model = whisper.load_model("medium")
|
19 |
+
# Get total audio duration in milliseconds
|
20 |
+
total_duration_ms = audio_segment.duration_seconds * 1000
|
21 |
+
|
22 |
+
# Calculate chunk duration in milliseconds
|
23 |
+
chunk_duration_ms = chunk_length_seconds * 1000
|
24 |
+
|
25 |
+
# Split audio into chunks
|
26 |
+
start_time = 0
|
27 |
+
chunk_num = 1
|
28 |
+
while start_time < total_duration_ms:
|
29 |
+
# Get the end time for the current chunk
|
30 |
+
end_time = min(start_time + chunk_duration_ms, total_duration_ms)
|
31 |
+
|
32 |
+
# Extract the current chunk
|
33 |
+
chunk = audio_segment[start_time:end_time]
|
34 |
+
|
35 |
+
# Generate output filename with sequential numbering
|
36 |
+
output_filename = f"{os.path.splitext(os.path.basename(audio_file))[0]}_chunk_{chunk_num}.{output_format}"
|
37 |
+
|
38 |
+
# Export the chunk as the specified format
|
39 |
+
chunk.export(output_filename, format=output_format)
|
40 |
+
|
41 |
+
# Update start time for the next chunk
|
42 |
+
start_time += chunk_duration_ms
|
43 |
+
chunk_num += 1
|
44 |
+
#transcribe Chunks
|
45 |
+
result = model.transcribe(output_filename)
|
46 |
+
#translate the transcribe data
|
47 |
+
translator=GoogleTranslator(source='auto',target=lang)
|
48 |
+
data_trans=translator.translate(result['text'])
|
49 |
+
Transcribe_Text.append(data_trans)
|
50 |
+
print(data_trans)
|
51 |
+
print(result['text'])
|
52 |
+
|
53 |
+
return Transcribe_Text
|
54 |
+
print(f"FLAC file '{flac_filepath}' successfully split into {chunk_num - 1} chunks.")
|
55 |
+
|
56 |
+
|