app.py

import gradio as gr
from transformers import pipeline

classifier = pipeline('audio-classification', model="steviebarot/filipino-SER")

def understand(audio):
    emotions = classifier(audio)
    labels = [i['label'] for i in emotions]
    scores = [str(round(i['score']*100, 2))+'%' for i in emotions]
    final_string = "\n".join([": ".join([l, s]) for l, s in zip(labels, scores)])
    return final_string

app = gr.Interface(fn=understand,
                   inputs=gr.Audio(source="microphone",
                                   type="filepath"),
                   outputs="text",
                   title="Filipino Speech Emotion Recognition",
                   description="Realtime demo for a Filipino Speech Emotion Recognition model. Created by Stevie Barot, Ken Agapito, and AJ Angcos.")
app.launch()