Update app.py

app.py

@@ -1,53 +1,56 @@
 from flask import Flask, render_template, request, jsonify
-import os
 import torch
-import speech_recognition as sr
 from transformers import pipeline
 from gtts import gTTS
+import os
 import re
 
 app = Flask(__name__)
-recognizer = sr.Recognizer()
 
-# Load Whisper Model (English Only)
+# Load Whisper Model for English Transcription
 device = "cuda" if torch.cuda.is_available() else "cpu"
-speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=0 if device == "cuda" else -1)
+asr_model = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=0 if device == "cuda" else -1)
 
-# Function to generate and save voice prompts in English
-def generate_audio(text, filename):
+# Function to generate audio prompts
+def generate_audio_prompt(text, filename):
     tts = gTTS(text=text, lang="en")
-    tts.save(filename)
+    tts.save(os.path.join("static", filename))
+
+# Generate audio prompts
+prompts = {
+    "welcome": "Welcome to Biryani Hub.",
+    "ask_name": "Tell me your name.",
+    "ask_email": "Please provide your email address.",
+    "thank_you": "Thank you for registration."
+}
 
-# Generate all voice prompts before starting
-generate_audio("Welcome to Biryani Hub.", "static/welcome.mp3")
-generate_audio("Tell me your name.", "static/ask_name.mp3")
-generate_audio("Please provide your email.", "static/ask_email.mp3")
-generate_audio("Thank you for registration.", "static/thank_you.mp3")
+for key, text in prompts.items():
+    generate_audio_prompt(text, f"{key}.mp3")
 
-# Function to clean text and remove non-English characters
-def clean_text(text):
-    return re.sub(r'[^a-zA-Z0-9@.\s]', '', text)  # Allow only English letters, numbers, @, and spaces
+# Clean transcribed text to allow only English letters, numbers, and basic punctuation
+def clean_transcription(text):
+    return re.sub(r"[^a-zA-Z0-9@.\s]", "", text)
 
 @app.route("/")
-def home():
+def index():
     return render_template("index.html")
 
-@app.route("/process_audio", methods=["POST"])
-def process_audio():
+@app.route("/transcribe", methods=["POST"])
+def transcribe():
     if "audio" not in request.files:
-        return jsonify({"error": "No audio file"}), 400
+        return jsonify({"error": "No audio file provided"}), 400
 
     audio_file = request.files["audio"]
-    audio_path = "static/temp.wav"
+    audio_path = os.path.join("static", "temp.wav")
     audio_file.save(audio_path)
 
     try:
-        # Force Whisper to transcribe in English only
-        text = speech_to_text(audio_path, generate_kwargs={"language": "en"})["text"]
-        cleaned_text = clean_text(text)  # Clean non-English characters
-        return jsonify({"text": cleaned_text})
+        # Transcribe audio to text
+        result = asr_model(audio_path, generate_kwargs={"language": "en"})
+        transcribed_text = clean_transcription(result["text"])
+        return jsonify({"text": transcribed_text})
     except Exception as e:
         return jsonify({"error": str(e)}), 500
 
 if __name__ == "__main__":
-    app.run(host="0.0.0.0", port=7860, debug=True)
+    app.run(host="0.0.0.0", port=5000, debug=True)