Update app.py

app.py

@@ -0,0 +1,93 @@
+import gradio as gr
+import requests
+import whisper
+import os
+import sounddevice as sd
+import wave
+import numpy as np
+from fastapi import FastAPI, UploadFile, File
+from fastapi.responses import FileResponse
+from gtts import gTTS
+import uvicorn
+
+app = FastAPI()
+
+# Load Whisper model
+model = whisper.load_model("base")
+
+# LibreTranslate API (Open Source)
+TRANSLATE_API_URL = "https://libretranslate.com/translate"
+
+# Record audio
+AUDIO_FILE = "recorded_audio.wav"
+OUTPUT_AUDIO = "output.mp3"
+
+def record_audio(duration=5, samplerate=44100):
+    print("Recording...")
+    audio_data = sd.rec(int(duration * samplerate), samplerate=samplerate, channels=2, dtype=np.int16)
+    sd.wait()
+    print("Recording Stopped")
+
+    with wave.open(AUDIO_FILE, "wb") as wf:
+        wf.setnchannels(2)
+        wf.setsampwidth(2)
+        wf.setframerate(samplerate)
+        wf.writeframes(audio_data.tobytes())
+
+    return AUDIO_FILE
+
+# Transcribe speech
+def transcribe_audio(file_path):
+    result = model.transcribe(file_path)
+    return result["text"]
+
+# Translate text
+def translate_text(text, source_lang, target_lang):
+    response = requests.post(TRANSLATE_API_URL, data={"q": text, "source": source_lang, "target": target_lang})
+    return response.json().get("translatedText", "Translation Error")
+
+# Convert text to speech
+def text_to_speech(text, lang):
+    tts = gTTS(text=text, lang=lang)
+    tts.save(OUTPUT_AUDIO)
+    return OUTPUT_AUDIO
+
+@app.post("/upload-audio/")
+async def upload_audio(file: UploadFile = File(...)):
+    file_path = f"temp/{file.filename}"
+    with open(file_path, "wb") as audio_file:
+        audio_file.write(await file.read())
+
+    detected_text = transcribe_audio(file_path)
+    source_lang = "ur" if any(char in detected_text for char in "اآبپتثجچحخ") else "ps"
+    target_lang = "ps" if source_lang == "ur" else "ur"
+
+    translated_text = translate_text(detected_text, source_lang, target_lang)
+    text_to_speech(translated_text, target_lang)
+
+    return {"text": detected_text, "translated_text": translated_text, "audio": "/get-audio/"}
+
+@app.get("/get-audio/")
+def get_audio():
+    return FileResponse(OUTPUT_AUDIO, media_type="audio/mpeg")
+
+# Gradio UI
+def process_audio():
+    files = {"file": open(AUDIO_FILE, "rb")}
+    response = requests.post("http://127.0.0.1:8000/upload-audio/", files=files).json()
+    return response["text"], response["translated_text"], response["audio"]
+
+with gr.Blocks() as demo:
+    gr.Markdown("### 🎙️ Urdu ↔ Pashto Voice Translator")
+    record_btn = gr.Button("🎤 Record")
+    stop_btn = gr.Button("⏹️ Stop")
+    translate_btn = gr.Button("🔄 Translate")
+    transcribed_text = gr.Textbox(label="Transcribed Text")
+    translated_text = gr.Textbox(label="Translated Text")
+    audio_output = gr.Audio(label="Translated Audio")
+
+    record_btn.click(fn=record_audio, inputs=[], outputs=[])
+    stop_btn.click(fn=lambda: None, inputs=[], outputs=[])
+    translate_btn.click(fn=process_audio, inputs=[], outputs=[transcribed_text, translated_text, audio_output])
+
+demo.launch()