AIVoice4

Sleeping

App Files Files Community

dschandra commited on Jan 2

Commit

cce0f39

verified ·

1 Parent(s): de42bf3

Update app.py

Browse files

Files changed (1) hide show

app.py +46 -9

app.py CHANGED Viewed

@@ -3,9 +3,16 @@ import speech_recognition as sr
 from tempfile import NamedTemporaryFile
 import os
 import ffmpeg
 app = Flask(__name__)
 html_code = """
 <!DOCTYPE html>
 <html lang="en">
@@ -60,14 +67,14 @@ html_code = """
         function startListening() {
             navigator.mediaDevices.getUserMedia({ audio: true }).then(stream => {
-                mediaRecorder = new MediaRecorder(stream);
                 mediaRecorder.start();
                 status.textContent = 'Listening...';
                 status.classList.add('listening');
                 audioChunks = [];
                 mediaRecorder.ondataavailable = event => audioChunks.push(event.data);
                 mediaRecorder.onstop = async () => {
-                    const audioBlob = new Blob(audioChunks, { type: 'audio/wav; codecs=LINEAR16' });
                     const formData = new FormData();
                     formData.append('audio', audioBlob);
@@ -116,30 +123,60 @@ def index():
 @app.route('/process-audio', methods=['POST'])
 def process_audio():
     try:
-        audio_file = request.files['audio']
-        temp_file = NamedTemporaryFile(delete=False, suffix=".wav")
         audio_file.save(temp_file.name)
         # Convert audio to PCM WAV format
         converted_file = NamedTemporaryFile(delete=False, suffix=".wav")
-        ffmpeg.input(temp_file.name).output(converted_file.name, acodec='pcm_s16le', ac=1, ar='16000').run(overwrite_output=True)
         recognizer = sr.Recognizer()
         with sr.AudioFile(converted_file.name) as source:
             audio_data = recognizer.record(source)
             command = recognizer.recognize_google(audio_data)
             response = process_command(command)
-        # Clean up temporary files
-        os.unlink(temp_file.name)
-        os.unlink(converted_file.name)
         return jsonify({"response": response})
     except Exception as e:
         return jsonify({"response": f"An error occurred: {str(e)}"})
 def process_command(command):
     """Process the user's voice command and return a response."""
-    global orders
     command = command.lower()
     if "menu" in command:
         return "Our menu includes paneer butter masala, fried rice, and cold coffee. What would you like to order?"

 from tempfile import NamedTemporaryFile
 import os
 import ffmpeg
+import logging
+from werkzeug.exceptions import BadRequest
+# Initialize Flask App
 app = Flask(__name__)
+# Set up logging
+logging.basicConfig(level=logging.INFO)
+# HTML Template for Frontend
 html_code = """
 <!DOCTYPE html>
 <html lang="en">
         function startListening() {
             navigator.mediaDevices.getUserMedia({ audio: true }).then(stream => {
+                mediaRecorder = new MediaRecorder(stream, { mimeType: 'audio/webm;codecs=opus' });
                 mediaRecorder.start();
                 status.textContent = 'Listening...';
                 status.classList.add('listening');
                 audioChunks = [];
                 mediaRecorder.ondataavailable = event => audioChunks.push(event.data);
                 mediaRecorder.onstop = async () => {
+                    const audioBlob = new Blob(audioChunks, { type: 'audio/webm' });
                     const formData = new FormData();
                     formData.append('audio', audioBlob);
 @app.route('/process-audio', methods=['POST'])
 def process_audio():
     try:
+        # Validate audio file
+        audio_file = request.files.get('audio')
+        if not audio_file:
+            raise BadRequest("No audio file provided.")
+        temp_file = NamedTemporaryFile(delete=False, suffix=".webm")
         audio_file.save(temp_file.name)
+        logging.info(f"Saved input audio to {temp_file.name}")
+        if os.path.getsize(temp_file.name) == 0:
+            raise BadRequest("Uploaded audio file is empty.")
         # Convert audio to PCM WAV format
         converted_file = NamedTemporaryFile(delete=False, suffix=".wav")
+        try:
+            ffmpeg.input(temp_file.name).output(
+                converted_file.name, acodec='pcm_s16le', ac=1, ar='16000'
+            ).run(overwrite_output=True)
+        except Exception as ffmpeg_error:
+            logging.error(f"FFmpeg conversion error: {str(ffmpeg_error)}")
+            return jsonify({"response": "Audio conversion failed. Please try again."})
+        logging.info(f"Converted audio saved to {converted_file.name}")
+        # Recognize speech
         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)
         return jsonify({"response": response})
+    except BadRequest as br:
+        logging.error(f"Bad request error: {br}")
+        return jsonify({"response": f"Bad Request: {str(br)}"})
     except Exception as e:
+        logging.error(f"Error processing audio: {e}")
         return jsonify({"response": f"An error occurred: {str(e)}"})
+    finally:
+        # Clean up temporary files
+        try:
+            if os.path.exists(temp_file.name):
+                os.unlink(temp_file.name)
+            if os.path.exists(converted_file.name):
+                os.unlink(converted_file.name)
+        except Exception as cleanup_error:
+            logging.error(f"Error cleaning up files: {cleanup_error}")
 def process_command(command):
     """Process the user's voice command and return a response."""
     command = command.lower()
     if "menu" in command:
         return "Our menu includes paneer butter masala, fried rice, and cold coffee. What would you like to order?"