app.py

import gradio as gr
from transformers import pipeline

trans=pipeline("automatic-speech-recognition",model="facebook/wav2vec2-large-xlsr-53-spanish")
clasificador=pipeline("text-classification",model="pysentimiento/robertuito-sentiment-analysis")

def audio2text(audio):
  text = trans(audio)["text"]
  return text

def text2sentiment(text):
  return clasificador(text)[0]["label"]

demo=gr.Blocks()
with demo:
    gr.Markdown("Audio 2 Text 2 Sentiment")
    audio=gr.Audio(sources="microphone",type="filepath")
    texto=gr.Textbox()
    b1=gr.Button("Transcribe")
    b1.click(audio2text,inputs=audio,outputs=texto)

    label=gr.Label()
    b2=gr.Button("Clasifica el sentimiento")
    b2.click(text2sentiment,inputs=texto,outputs=label)

demo.launch()