Update app.py

app.py

@@ -1,59 +1,69 @@
-import streamlit as st
-import sounddevice as sd
-import numpy as np
-import wavio
+from flask import Flask, request, jsonify, send_file
 import speech_recognition as sr
-from transformers import pipeline, AutoModelForSeq2SeqLM, AutoTokenizer
+from googletrans import Translator
 from gtts import gTTS
 import tempfile
 import os
 
-# Load translation model
-model_name = "Helsinki-NLP/opus-mt-ur-en"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
-translator = pipeline("translation", model=model, tokenizer=tokenizer)
-
-def record_audio(duration=5, samplerate=44100):
-    st.write("🎤 Recording...")
-    audio = sd.rec(int(duration * samplerate), samplerate=samplerate, channels=2, dtype=np.int16)
-    sd.wait()
-    temp_audio_path = "recorded_audio.wav"
-    wavio.write(temp_audio_path, audio, samplerate, sampwidth=2)
-    return temp_audio_path
+app = Flask(__name__)
+translator = Translator()
 
+# Function to recognize speech from an audio file
 def recognize_speech(audio_file):
     recognizer = sr.Recognizer()
     with sr.AudioFile(audio_file) as source:
-        audio_data = recognizer.record(source)
-        try:
-            text = recognizer.recognize_google(audio_data, language="ur-PK")
-            return text
-        except sr.UnknownValueError:
-            return "Could not understand audio"
-        except sr.RequestError:
-            return "Could not request results"
+        audio = recognizer.record(source)
+    try:
+        return recognizer.recognize_google(audio, language="ps")  # Default language: Pashto
+    except sr.UnknownValueError:
+        return "Speech not recognized"
+    except sr.RequestError:
+        return "Speech recognition service unavailable"
+
+# Function to translate text using googletrans
+def translate_text(text, source_lang, target_lang):
+    translation = translator.translate(text, src=source_lang, dest=target_lang)
+    return translation.text
 
+# Function to convert text to speech
 def text_to_speech(text, lang):
-    tts = gTTS(text=text, lang=lang)
-    temp_audio = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
-    temp_audio.close()
-    tts.save(temp_audio.name)
-    return temp_audio.name
+    tts = gTTS(text, lang=lang)
+    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
+    tts.save(temp_file.name)
+    return temp_file.name
 
-st.title("🎙️ Voice-to-Voice Translator")
+# Route to handle voice input and translation
+@app.route("/translate", methods=["POST"])
+def translate():
+    if "audio" not in request.files:
+        return jsonify({"error": "No audio file uploaded"}), 400
 
-if st.button("🎤 Hold to Speak", key="record"):
-    audio_path = record_audio(duration=5)
-    recognized_text = recognize_speech(audio_path)
+    audio_file = request.files["audio"]
     
-    if "پښتو" in recognized_text or "Pashto" in recognized_text:
-        response = translator(recognized_text)[0]["translation_text"]
-        response_audio = text_to_speech(response, "ur")
-    else:
-        response = translator(recognized_text)[0]["translation_text"]
-        response_audio = text_to_speech(response, "ps")
+    # Save audio file temporarily
+    temp_audio = tempfile.NamedTemporaryFile(delete=False, suffix=".wav")
+    audio_file.save(temp_audio.name)
     
-    st.write("**Recognized Text:**", recognized_text)
-    st.write("**Translation:**", response)
-    st.audio(response_audio, format='audio/mp3')
+    # Recognize speech
+    text = recognize_speech(temp_audio.name)
+    os.remove(temp_audio.name)  # Delete after processing
+
+    if text == "Speech not recognized" or text == "Speech recognition service unavailable":
+        return jsonify({"error": text}), 500
+
+    # Translate from Pashto to Urdu
+    translated_text = translate_text(text, "ps", "ur")
+
+    # Convert to speech
+    audio_path = text_to_speech(translated_text, "ur")
+
+    return jsonify({"original_text": text, "translated_text": translated_text, "audio_path": audio_path})
+
+# Route to fetch the generated audio file
+@app.route("/audio", methods=["GET"])
+def get_audio():
+    audio_path = request.args.get("file")
+    return send_file(audio_path, mimetype="audio/mp3", as_attachment=True)
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=7860, debug=True)