Update app.py

app.py

@@ -14,30 +14,30 @@ if not os.path.exists('audio'):
 # Step 1: Transcribe the audio file
 def transcribe_audio(audio):
     if audio is None:
-        return "No audio file provided."  # Handle the case when no audio is uploaded
+        return "No audio file provided."
     
     recognizer = sr.Recognizer()
+    
+    # Check if the file exists
+    if not os.path.isfile(audio):
+        return "Audio file not found."
+
     audio_format = audio.split('.')[-1].lower()
     
-    # Convert to WAV if the audio is not in a supported format
     if audio_format != 'wav':
         try:
-            # Load the audio file with pydub
             audio_segment = AudioSegment.from_file(audio)
             wav_path = audio.replace(audio_format, 'wav')
-            audio_segment.export(wav_path, format='wav')  # Convert to WAV
-            audio = wav_path  # Update audio path to the converted file
+            audio_segment.export(wav_path, format='wav')
+            audio = wav_path
         except Exception as e:
             return f"Error converting audio: {e}"
 
-    # Convert audio into recognizable format for the Recognizer
     audio_file = sr.AudioFile(audio)
-    
     with audio_file as source:
         audio_data = recognizer.record(source)
 
     try:
-        # Recognize the audio using Google Web Speech API
         transcription = recognizer.recognize_google(audio_data)
         return transcription
     except sr.UnknownValueError:
@@ -45,38 +45,48 @@ def transcribe_audio(audio):
     except sr.RequestError as e:
         return f"Error with Google Speech Recognition service: {e}"
 
+
 # Step 2: Create pronunciation audio for incorrect words
 def upfilepath(local_filename):
-
-    # URL để upload tệp âm thanh
     upload_url = "https://mr2along-speech-recognize.hf.space/gradio_api/upload?upload_id=yw08d344te"
-    # Dữ liệu tệp cần upload
     files = {'files': open(local_filename, 'rb')}
-    
-    # Gửi yêu cầu POST
-    response = requests.post(upload_url, files=files)
-    
-    # Kiểm tra kết quả trả về từ server
-    if response.status_code == 200:
-        print("Upload thành công!")
-        result=response.json()
-        extracted_path = result[0]
-        print(extracted_path)  # In kết quả nếu server trả về dưới dạng JSON
-        return extracted_path
-    else:
-        print(f"Lỗi: {response.status_code}")
-        print(response.text)  # In thông báo lỗi từ server
+
+    try:
+        response = requests.post(upload_url, files=files, timeout=30)  # Set timeout (e.g., 30 seconds)
+        
+        if response.status_code == 200:
+            print("Upload thành công!")
+            result = response.json()
+            extracted_path = result[0]
+            print(extracted_path)
+            return extracted_path
+        else:
+            print(f"Lỗi: {response.status_code}")
+            print(response.text)
+            return None
+
+    except requests.exceptions.Timeout:
+        return "Request timed out. Please try again."
+    except Exception as e:
+        return f"An error occurred: {e}"
+
         
 def create_pronunciation_audio(word):
-    time.sleep(1)  # Chờ 5 giây
-    tts = gTTS(word)
-    main_url="https://mr2along-speech-recognize.hf.space/gradio_api/file="
-    
-    audio_file_path = f"audio/{word}.mp3"  # Save the audio to a file
-    tts.save(audio_file_path)
-    word_audio=upfilepath(audio_file_path)
-    #print(f"Lỗi: {word_audio}")
-    return f"{main_url}{word_audio}"  # Return the file path of the saved audio
+    retries = 3  # Retry up to 3 times
+    for attempt in range(retries):
+        try:
+            tts = gTTS(word)
+            audio_file_path = f"audio/{word}.mp3"
+            tts.save(audio_file_path)
+            word_audio = upfilepath(audio_file_path)
+            if word_audio:
+                return f"https://mr2along-speech-recognize.hf.space/gradio_api/file={word_audio}"
+        except Exception as e:
+            if attempt < retries - 1:
+                time.sleep(2 ** attempt)  # Exponential backoff
+            else:
+                return f"Failed to create pronunciation audio: {e}"
+
 
 # Step 3: Compare the transcribed text with the input paragraph
 def compare_texts(reference_text, transcribed_text):