Update app.py

app.py

@@ -1,63 +1,29 @@
-import whisper
-from transformers import MarianMTModel, MarianTokenizer
-from gtts import gTTS
-import tempfile
-import os
-import certifi
-import gradio as gr
 
-os.environ["SSL_CERT_FILE"] = certifi.where()
+import gradio as gr
+from processing import transcribe_translate_speak
+import tempfile
 
-def process_audio(audio_file, target_language):
-    # Save audio to temp file
+def process(audio_file, language):
     with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp:
         tmp.write(audio_file.read())
         tmp_path = tmp.name
 
-    # 1. Transcribe with Whisper
-    whisper_model = whisper.load_model("tiny")
-    result = whisper_model.transcribe(tmp_path)
-    text = result["text"]
-
-    # 2. Translate text
-    lang_map = {
-        'hi': "Helsinki-NLP/opus-mt-en-hi",
-        'es': "Helsinki-NLP/opus-mt-en-es",
-        'fr': "Helsinki-NLP/opus-mt-en-fr",
-        'bn': "shhossain/opus-mt-en-to-bn"
-    }
-
-    if target_language not in lang_map:
-        return "Unsupported language selected", None
-
-    model_name = lang_map[target_language]
-    tokenizer = MarianTokenizer.from_pretrained(model_name)
-    translation_model = MarianMTModel.from_pretrained(model_name)
-
-    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
-    outputs = translation_model.generate(**inputs)
-    translated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-
-    # 3. Convert to speech with gTTS
-    tts = gTTS(translated_text, lang=target_language)
-    output_path = tempfile.NamedTemporaryFile(suffix=".mp3", delete=False).name
-    tts.save(output_path)
-
-    return translated_text, output_path
+    transcription, translation, audio_output = transcribe_translate_speak(tmp_path, language)
+    return transcription, translation, audio_output
 
-# Gradio Interface
 iface = gr.Interface(
-    fn=process_audio,
+    fn=process,
     inputs=[
-        gr.Audio(source="upload", type="file", label="Upload English Audio"),
-        gr.Dropdown(["hi", "es", "fr", "bn"], label="Target Language")
+        gr.Audio(source="microphone", type="file", label="Speak or Upload Audio"),
+        gr.Dropdown(["hi", "es", "fr", "bn"], label="Translate To")
     ],
     outputs=[
+        gr.Textbox(label="Transcribed Text"),
         gr.Textbox(label="Translated Text"),
-        gr.Audio(label="Output Audio")
+        gr.Audio(label="Translated Audio")
     ],
-    title="Audio Translator",
-    description="Upload an English audio clip to transcribe it, translate it to another language, and hear it spoken."
+    title="🎙️ SpeechSync - Python Edition",
+    description="Record your voice, and this app will transcribe, translate, and speak it in your chosen language."
 )
 
 if __name__ == "__main__":