Spaces:
Sleeping
Sleeping
File size: 3,950 Bytes
3c02b94 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 | # import whisper
# import tempfile
# import os
# class AudioProcessor:
# def __init__(self):
# self.model = whisper.load_model("base")
# def transcribe(self, audio_file):
# with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as tmp:
# tmp.write(audio_file.read())
# tmp_path = tmp.name
# try:
# result = self.model.transcribe(tmp_path)
# text = result['text']
# segments = result.get('segments', [])
# avg_confidence = 0.0
# if segments:
# confidences = [s.get('no_speech_prob', 0) for s in segments]
# avg_confidence = 1 - (sum(confidences) / len(confidences))
# else:
# avg_confidence = 0.8
# return {
# 'text': text,
# 'confidence': avg_confidence,
# 'needs_review': avg_confidence < 0.6
# }
# finally:
# os.unlink(tmp_path)
# import whisper
# import tempfile
# import os
# import soundfile as sf
# import numpy as np
# class AudioProcessor:
# def __init__(self):
# self.model = whisper.load_model("base")
# def transcribe(self, audio_file):
# # Save uploaded file
# with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as tmp:
# tmp.write(audio_file.read())
# tmp_path = tmp.name
# try:
# # Transcribe directly
# result = self.model.transcribe(tmp_path, fp16=False)
# text = result['text']
# segments = result.get('segments', [])
# avg_confidence = 0.8
# if segments:
# confidences = [1 - s.get('no_speech_prob', 0) for s in segments]
# avg_confidence = sum(confidences) / len(confidences)
# return {
# 'text': text,
# 'confidence': avg_confidence,
# 'needs_review': avg_confidence < 0.6
# }
# except Exception as e:
# # Fallback
# return {
# 'text': "Error transcribing audio. Please try again.",
# 'confidence': 0.0,
# 'needs_review': True
# }
# finally:
# if os.path.exists(tmp_path):
# os.unlink(tmp_path)
import whisper
import os
class AudioProcessor:
def __init__(self):
try:
self.model = whisper.load_model("tiny") # Smaller, faster
except:
self.model = None
def transcribe(self, audio_file):
if self.model is None:
return {
'text': "",
'confidence': 0.0,
'needs_review': True
}
# Save to temp file
temp_path = "temp_audio.wav"
try:
with open(temp_path, "wb") as f:
f.write(audio_file.getvalue())
# Transcribe
result = self.model.transcribe(temp_path, language="en", fp16=False)
text = result.get('text', '').strip()
return {
'text': text,
'confidence': 0.8,
'needs_review': len(text) < 5
}
except Exception as e:
print(f"Transcription error: {e}")
return {
'text': "",
'confidence': 0.0,
'needs_review': True
}
finally:
if os.path.exists(temp_path):
try:
os.remove(temp_path)
except:
pass |