Ichcanziho's picture
Create app.py
f96a028
from transformers import pipeline
import gradio as gr
from PIL import Image
def classify_img(im):
im = Image.fromarray(im.astype('uint8'), 'RGB')
ans = image_cla(im)
labels = {v["label"]: v["score"] for v in ans}
return labels
def voice2text(audio):
text = voice_cla(audio)["text"]
return text
def text2sentiment(text):
sentiment = text_cla(text)[0]["label"]
return sentiment
def make_block(dem):
with dem:
gr.Markdown("""
# Ejemplo de `space` multiclassifier:
Este `space` contiene los siguientes modelos:
- ASR: [Wav2Vec2](https://huggingface.co/facebook/wav2vec2-large-xlsr-53-spanish)
- Text Classification: [Robertuito](https://huggingface.co/pysentimiento/robertuito-sentiment-analysis)
- Image classifier: [Swin-small-patch4](https://huggingface.co/microsoft/swin-small-patch4-window7-224)
Autor del demo: [Gabriel Ichcanziho](https://www.linkedin.com/in/ichcanziho/)
Puedes probar un demo de cada uno en las siguientes pestañas:
""")
with gr.Tabs():
with gr.TabItem("Transcribe audio en español"):
with gr.Row():
audio = gr.Audio(source="microphone", type="filepath")
transcripcion = gr.Textbox()
b1 = gr.Button("Voz a Texto")
with gr.TabItem("Análisis de sentimiento en español"):
with gr.Row():
texto = gr.Textbox()
label = gr.Label()
b2 = gr.Button("Texto a Sentimiento")
with gr.TabItem("Clasificación de Imágenes"):
with gr.Row():
image = gr.Image(label="Carga una imagen aquí")
label_image = gr.Label(num_top_classes=5)
b3 = gr.Button("Clasifica")
b1.click(voice2text, inputs=audio, outputs=transcripcion)
b2.click(text2sentiment, inputs=texto, outputs=label)
b3.click(classify_img, inputs=image, outputs=label_image)
if __name__ == '__main__':
image_cla = pipeline("image-classification", model="microsoft/swin-tiny-patch4-window7-224")
voice_cla = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-xlsr-53-spanish")
text_cla = pipeline("text-classification", model="pysentimiento/robertuito-sentiment-analysis")
demo = gr.Blocks()
make_block(demo)
demo.launch()