Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -2,6 +2,18 @@ import gradio as gr
|
|
2 |
from transformers import pipeline
|
3 |
import torch
|
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
# Cargar el modelo que convierte imagen a texto
|
6 |
image_to_text_model = pipeline("image-classification")
|
7 |
|
@@ -11,13 +23,14 @@ text_to_audio_model = pipeline("text-to-speech")
|
|
11 |
# Funci贸n para la interfaz de Gradio
|
12 |
def image_to_audio(input_image):
|
13 |
# Convertir la imagen a texto
|
14 |
-
text_output = image_to_text_model(input_image)[0]['label']
|
15 |
|
16 |
# Generar audio a partir del texto
|
17 |
audio_output = text_to_audio_model(text_output)[0]['audio']
|
18 |
|
19 |
return audio_output
|
20 |
|
|
|
21 |
# Interfaz Gradio
|
22 |
iface = gr.Interface(
|
23 |
fn=image_to_audio,
|
@@ -29,7 +42,4 @@ iface = gr.Interface(
|
|
29 |
)
|
30 |
|
31 |
# Ejecutar la interfaz
|
32 |
-
iface.launch()
|
33 |
-
# Crear interfaz de Gradio
|
34 |
-
#iface = gr.Interface(fn=asr, inputs=gr.inputs.Audio(source="microphone", type="file"), outputs="text")
|
35 |
-
#iface.launch()
|
|
|
2 |
from transformers import pipeline
|
3 |
import torch
|
4 |
|
5 |
+
#Definir 2 modelos uno de imagen a texto y otro de texto a audio que inyecta
|
6 |
+
# el resultado del primero modelo(texto generado) en la entrada del 2潞 modelo
|
7 |
+
# texto to audio
|
8 |
+
|
9 |
+
def transform(example_batch):
|
10 |
+
# Take a list of PIL images and turn them to pixel values
|
11 |
+
inputs = feature_extractor([x.convert("RGB") for x in example_batch['image']], return_tensors='pt')
|
12 |
+
|
13 |
+
# Don't forget to include the labels!
|
14 |
+
inputs['labels'] = example_batch['labels']
|
15 |
+
return inputs
|
16 |
+
|
17 |
# Cargar el modelo que convierte imagen a texto
|
18 |
image_to_text_model = pipeline("image-classification")
|
19 |
|
|
|
23 |
# Funci贸n para la interfaz de Gradio
|
24 |
def image_to_audio(input_image):
|
25 |
# Convertir la imagen a texto
|
26 |
+
text_output = transform(image_to_text_model(input_image)[0]['label'])
|
27 |
|
28 |
# Generar audio a partir del texto
|
29 |
audio_output = text_to_audio_model(text_output)[0]['audio']
|
30 |
|
31 |
return audio_output
|
32 |
|
33 |
+
|
34 |
# Interfaz Gradio
|
35 |
iface = gr.Interface(
|
36 |
fn=image_to_audio,
|
|
|
42 |
)
|
43 |
|
44 |
# Ejecutar la interfaz
|
45 |
+
iface.launch()
|
|
|
|
|
|