app.py

import gradio as gr
from transformers import pipeline

# Load the Automatic Speech Recognition (ASR) model for Finnish
asr_model = pipeline("automatic-speech-recognition", model="openai/whisper-small")

# Load the translation model for Finnish to English
translator = pipeline("translation", model="Helsinki-NLP/opus-mt-fi-en")

# Function to handle translation from Finnish speech to English text
def translate_speech(audio):
    # Convert Finnish speech to Finnish text
    finnish_text = asr_model(audio)["text"]
    # Translate Finnish text to English text
    english_text = translator(finnish_text)[0]["translation_text"]
    return english_text

# Build Gradio Interface
interface = gr.Interface(
    fn=translate_speech,
    inputs=gr.Audio(source=["microphone", "upload"], type="filepath"),
    outputs="text",
    title="Finnish to English Speech Translator",
    description="This app translates Finnish speech to English text. You can either speak in Finnish or upload an audio file, and see the translation appear in English!"
)

# Launch the app
interface.launch(api=True)