from transformers import pipeline import gradio as gr from song_guesser import SongGuesser pipe = pipeline(model="MarieGotthardt/whisper") # change to "your-username/the-name-you-picked" def transcribe(audio): text = pipe(audio)["text"] return text def get_song_guess(audio): user_query = transcribe(audio) output = "Vad vi hörde från dig: " + user_query + "\n\nVi tror att du sjöng: " + SongGuesser.guess_song(user_query) return output iface = gr.Interface( fn=get_song_guess, inputs=gr.Audio(sources=["microphone"], type="filepath"), outputs="text", title="Sjung en sång och låt oss gissa 🎤 🎄", description="Sing a Swedish Christmas song and see if we can guess it. If our guess is wrong, try singing more clearly, more of the lyrics, or another song and maybe we will get it the next time!", ) iface.launch()