Commit
•
4c13691
1
Parent(s):
99b3be8
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import torch
|
2 |
+
import gradio as gr
|
3 |
+
from PIL import Image
|
4 |
+
import scipy.io.wavfile as wavfile
|
5 |
+
|
6 |
+
# Use a pipeline as a high-level helper
|
7 |
+
from transformers import pipeline
|
8 |
+
|
9 |
+
# from phonemizer.backend.espeak.wrapper import EspeakWrapper
|
10 |
+
|
11 |
+
# _ESPEAK_LIBRARY = '/opt/homebrew/Cellar/espeak/1.48.04_1/lib/libespeak.1.1.48.dylib' #use the Path to the library.
|
12 |
+
# EspeakWrapper.set_library(_ESPEAK_LIBRARY)
|
13 |
+
|
14 |
+
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
|
15 |
+
|
16 |
+
narrator = pipeline("text-to-speech", model="kakao-enterprise/vits-ljs")
|
17 |
+
|
18 |
+
# tts_model_path = "./Model/models--kakao-enterprise--vits-ljs/snapshots/3bcb8321394f671bd948ebf0d086d694dda95464"
|
19 |
+
|
20 |
+
# narrator = pipeline("text-to-speech", model=tts_model_path)
|
21 |
+
|
22 |
+
# Load the pretrained weights
|
23 |
+
caption_image = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large", device=device)
|
24 |
+
|
25 |
+
# model_path = "./Model/models--Salesforce--blip-image-captioning-large/snapshots/2227ac38c9f16105cb0412e7cab4759978a8fd90"
|
26 |
+
|
27 |
+
# Load the pretrained weights
|
28 |
+
# caption_image = pipeline("image-to-text", model=model_path, device=device)
|
29 |
+
|
30 |
+
# define the function to generate audio from text
|
31 |
+
def generate_audio(text):
|
32 |
+
|
33 |
+
# generate the narrated text
|
34 |
+
narrated_text = narrator(text)
|
35 |
+
|
36 |
+
# save the audio to WAV file
|
37 |
+
wavfile.write("output.wav", rate=narrated_text["sampling_rate"],
|
38 |
+
data=narrated_text["audio"][0])
|
39 |
+
|
40 |
+
# Return the path to the saved output WAV file
|
41 |
+
return "output.wav"
|
42 |
+
|
43 |
+
|
44 |
+
def caption_my_image(pil_image):
|
45 |
+
|
46 |
+
semantics = caption_image(pil_image)[0]["generated_text"]
|
47 |
+
audio = generate_audio(semantics)
|
48 |
+
|
49 |
+
return audio
|
50 |
+
|
51 |
+
|
52 |
+
gr.close_all()
|
53 |
+
|
54 |
+
demo = gr.Interface(fn=caption_my_image,
|
55 |
+
inputs=[gr.Image(label="Select Image", type="pil")],
|
56 |
+
outputs=[gr.Audio(label="Generated Audio")],
|
57 |
+
title="@IT AI Enthusiast (https://www.youtube.com/@itaienthusiast/) - Project 8: Image Captioning with AI",
|
58 |
+
description="THIS APPLICATION WILL BE USED TO CAPTION IMAGES WITH THE HELP OF AI")
|
59 |
+
|
60 |
+
demo.launch()
|