Update app.py

app.py

@@ -85,19 +85,16 @@ html_code = """
         let mediaRecorder;
         let audioChunks = [];
         let isConversationActive = false;
-
         micButton.addEventListener('click', () => {
             if (!isConversationActive) {
                 isConversationActive = true;
                 startConversation();
             }
         });
-
         function startConversation() {
             status.textContent = 'Listening...';
             startListening();
         }
-
         function startListening() {
             navigator.mediaDevices.getUserMedia({ audio: true }).then(stream => {
                 mediaRecorder = new MediaRecorder(stream, { mimeType: 'audio/webm;codecs=opus' });
@@ -108,22 +105,18 @@ html_code = """
                     const audioBlob = new Blob(audioChunks, { type: 'audio/webm' });
                     const formData = new FormData();
                     formData.append('audio', audioBlob);
-
                     status.textContent = 'Processing...';
                     try {
                         const result = await fetch('/process-audio', { method: 'POST', body: formData });
                         const data = await result.json();
                         response.textContent = data.response;
                         response.style.display = 'block';
-
                         try {
                             const utterance = new SpeechSynthesisUtterance(data.response);
                             speechSynthesis.speak(utterance);
-
                             utterance.onend = () => {
                                 console.log("Speech synthesis completed.");
                             };
-
                             utterance.onerror = (e) => {
                                 console.error("Speech synthesis error:", e.error);
                                 status.textContent = 'Error with speech output.';
@@ -132,7 +125,6 @@ html_code = """
                             console.error("Speech synthesis not supported or failed:", speechError);
                             response.textContent = "Speech output unavailable. Please check your browser.";
                         }
-
                         if (data.response.includes("Goodbye")) {
                             status.textContent = 'Conversation ended. Press the mic button to start again.';
                             isConversationActive = false;
@@ -177,7 +169,7 @@ def process_audio():
         if os.path.getsize(temp_file.name) == 0:
             raise BadRequest("Uploaded audio file is empty.")
 
-        # Convert audio to PCM WAV format
+        # Convert audio to PCM WAV format (16kHz, mono)
         converted_file = NamedTemporaryFile(delete=False, suffix=".wav")
         try:
             ffmpeg.input(temp_file.name).output(
@@ -193,9 +185,16 @@ def process_audio():
         recognizer = sr.Recognizer()
         with sr.AudioFile(converted_file.name) as source:
             audio_data = recognizer.record(source)
-            command = recognizer.recognize_google(audio_data)
-            logging.info(f"Recognized command: {command}")
-            response = process_command(command)
+            try:
+                command = recognizer.recognize_google(audio_data)
+                logging.info(f"Recognized command: {command}")
+                response = process_command(command)
+            except sr.UnknownValueError:
+                logging.error("Google Speech Recognition could not understand the audio")
+                response = "Sorry, I couldn't understand your request. Please try again."
+            except sr.RequestError as e:
+                logging.error(f"Error with Google Speech Recognition service: {e}")
+                response = "Sorry, there was an issue with the speech recognition service."
 
         return jsonify({"response": response})