from pytube import YouTube
from transformers import pipeline
import gradio as gr
import os



pipe = pipeline(model = "Valdimarb13/whisper-small-icelandic") 

def transcribe(audio):
    text = pipe(audio)["text"]
    return text

def get_audio(url):
  yt = YouTube(url)
  video = yt.streams.filter(only_audio=True).first()
  out_file=video.download(output_path=".")
  base, ext = os.path.splitext(out_file)
  new_file = base+'.mp3'
  os.rename(out_file, new_file)
  a = new_file
  return a


def get_text(url):
  text = pipe(get_audio(url))["text"]
  return text


with gr.Blocks() as demo:
    with gr.Tab('Enter the URL of a video with Icelandic speech to get a transcription'):
        with gr.Row():
            input_text = gr.Textbox(placeholder='Enter the Youtube video URL', label='URL')
            output_text = gr.Textbox(placeholder='Transcription of the video', label='Transcription')
        youtube_button = gr.Button('Get Transcription')
    with gr.Tab("Speak into the microphone to get a transcription"):
        with gr.Row():
            input_speech = gr.Audio(source="microphone", type="filepath")
            output_speech =  gr.Textbox("Transcribed audio")
        speech_button = gr.Button("Submit")
        
    
    youtube_button.click(get_text, inputs = input_text, outputs = output_text)
    speech_button.click(transcribe, inputs = input_speech, outputs = output_speech)

demo.launch()