first commit

app.py

@@ -0,0 +1,108 @@
+from flask import Flask, render_template, request, jsonify
+import requests
+import json
+import speech_recognition as sr
+import io
+import base64
+import os
+
+app = Flask(__name__)
+
+SUPPORTED_LANGUAGES = {
+    "arabic": "Arabic",
+    "armenian": "Armenian",
+    "azerbaijani": "Azerbaijani",
+    "english": "English",
+    "french": "French",
+    "georgian": "Georgian",
+    "kazakh": "Kazakh",
+    "mandarin": "Mandarin Chinese",
+    "persian": "Persian (Farsi)",
+    "portuguese": "Portuguese",
+    "russian": "Russian",
+    "turkish": "Turkish",
+    "uzbek": "Uzbek"
+}
+
+OLLAMA_API_URL = "http://100.64.0.4:11434/api/generate"
+
+@app.route('/')
+def index():
+    return render_template('index.html', languages=SUPPORTED_LANGUAGES)
+
+@app.route('/translate', methods=['POST'])
+def translate():
+    data = request.json
+    source_language = data.get('sourceLanguage')
+    target_language = data.get('targetLanguage')
+    text = data.get('text')
+
+    if not all([source_language, target_language, text]):
+        return jsonify({"error": "Missing required parameters"}), 400
+
+    # Create prompt for Gemma 3
+    prompt = f"""Translate the following text from {source_language} to {target_language}.
+Text to translate: {text}
+
+Provide ONLY the translated text with no additional commentary, explanations, or formatting.
+"""
+
+    # Call Ollama API with the Gemma 3 model
+    payload = {
+        "model": "gemma3:12b",
+        "prompt": prompt,
+        "stream": False
+    }
+    
+    try:
+        response = requests.post(OLLAMA_API_URL, json=payload)
+        response.raise_for_status()
+        result = response.json()
+        
+        # Extract the generated translation
+        translation = result.get("response", "").strip()
+        
+        return jsonify({"translation": translation})
+    except Exception as e:
+        return jsonify({"error": f"Translation failed: {str(e)}"}), 500
+
+@app.route('/speech-to-text', methods=['POST'])
+def speech_to_text():
+    try:
+        audio_data = request.json.get('audio')
+        language_code = request.json.get('language')
+        
+        # Convert base64 audio to file
+        audio_bytes = base64.b64decode(audio_data.split(',')[1])
+        
+        # Use speech recognition
+        recognizer = sr.Recognizer()
+        with sr.AudioFile(io.BytesIO(audio_bytes)) as source:
+            audio = recognizer.record(source)
+            
+        # Convert speech to text
+        language_code_map = {
+            "arabic": "ar-AR",
+            "armenian": "hy-AM",
+            "azerbaijani": "az-AZ",
+            "english": "en-US",
+            "french": "fr-FR",
+            "georgian": "ka-GE",
+            "kazakh": "kk-KZ",
+            "mandarin": "zh-CN",
+            "persian": "fa-IR",
+            "portuguese": "pt-PT",
+            "russian": "ru-RU", 
+            "turkish": "tr-TR",
+            "uzbek": "uz-UZ"
+        }
+        
+        lang_code = language_code_map.get(language_code, "en-US")
+        text = recognizer.recognize_google(audio, language=lang_code)
+        
+        return jsonify({"text": text})
+    except Exception as e:
+        return jsonify({"error": f"Speech recognition failed: {str(e)}"}), 500
+
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=5005, debug=True)