app.py

import gradio as gr
import whisper
from deep_translator import GoogleTranslator
from gtts import gTTS
import tempfile
import os
import warnings
warnings.filterwarnings("ignore")

model = whisper.load_model("base")

def translate_audio(text_input, upload_audio, mic_audio, source_lang="tr", target_lang="en"):
    audio_file = None
    audio_source = ""
    
    if mic_audio is not None and mic_audio != "":
        audio_file = mic_audio
        audio_source = "Microphone"
        print(f"Using microphone audio: {mic_audio}")
    elif upload_audio is not None:
        audio_file = upload_audio
        audio_source = "Upload"
        print(f"Using uploaded audio: {upload_audio}")
    elif text_input and text_input.strip():
        try:
            original_text = text_input.strip()
            
            if source_lang != target_lang:
                translator = GoogleTranslator(source=source_lang, target=target_lang)
                translated_text = translator.translate(original_text)
            else:
                translated_text = original_text
            
            tts = gTTS(text=translated_text, lang=target_lang, slow=False)
            
            with tempfile.NamedTemporaryFile(delete=False, suffix='.mp3') as tmp_file:
                tts.save(tmp_file.name)
                return (
                    tmp_file.name, 
                    f"Original: {original_text}", 
                    f"Translated: {translated_text}",
                    f" Source:  Text Input"
                )
        except Exception as e:
            return None, f" Error: {str(e)}", "Please try again", "Source:  Text Input"
    else:
        return None, " No input provided", "Please upload audio, record with microphone, OR enter text", ""
    
    try:
        print(f"Processing {audio_source} audio: {audio_file}")
        
        result = model.transcribe(audio_file, language=source_lang, fp16=False)
        original_text = result["text"].strip()
        
        if not original_text:
            return None, " No speech detected", "Please try again with clearer audio", f"Source: {audio_source}"
        
        print(f"Transcribed: {original_text}")
        
        if source_lang != target_lang:
            translator = GoogleTranslator(source=source_lang, target=target_lang)
            translated_text = translator.translate(original_text)
        else:
            translated_text = original_text
        
        print(f"Translated: {translated_text}")
        
        tts = gTTS(text=translated_text, lang=target_lang, slow=False)
        
        with tempfile.NamedTemporaryFile(delete=False, suffix='.mp3') as tmp_file:
            tts.save(tmp_file.name)
            return (
                tmp_file.name, 
                f"Original: {original_text}", 
                f" Translated: {translated_text}",
                f" Source: {audio_source}"
            )
            
    except Exception as e:
        print(f"Error: {str(e)}")
        return None, f" Error: {str(e)}", "Please try again", f"Source: {audio_source}"

languages = {
    "tr": "Turkish",
    "en": "English", 
    "fr": "French",
    "es": "Spanish",
    "de": "German",
    "hi": "Hindi",
    "ja": "Japanese",
    "ar": "Arabic",
    "ru": "Russian",
    "zh": "Chinese"
}

demo = gr.Interface(
    fn=translate_audio,
    inputs=[
        gr.Textbox(
            label="Enter text to translate",
            placeholder="Type text here or leave empty to use audio input...",
            lines=2
        ),
        gr.Audio(
            type="filepath",         
            sources=["upload"],
            label="Upload Audio File"
        ),
        gr.Audio(
            sources=["microphone"], 
            label="OR Record with Microphone",
            type="filepath"
        ),
        gr.Dropdown(
            choices=list(languages.keys()),
            value="tr",
            label="Source Language"
        ),
        gr.Dropdown(
            choices=list(languages.keys()),
            value="en", 
            label="Target Language"
        )
    ],
    outputs=[
        gr.Audio(label="Translated Audio Output"),
        gr.Textbox(label="Original Text"),
        gr.Textbox(label="Translated Text"),
        gr.Textbox(label="Audio Source Info")
    ],
    
    title="VoiceAI61",
    article="""
    <div style="text-align: center; margin-top: 20px;">
        <p>📱 <strong>dear users:</strong> trabzon champion. </p>
    </div>
    """,
    
    examples=[
        ["", None, None, "tr", "en"],
        ["", None, None, "en", "fr"],
        ["Hello world", None, None, "en", "es"],
    ],
    
    allow_flagging="never"
)

if __name__ == "__main__":
    demo.launch()