Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# prompt: add the theme featrue in above Audio Transcription, Translation, and Sentiment Analysis app and also need selectbox for sentiment and generated image
|
2 |
+
import whisper
|
3 |
+
import os
|
4 |
+
import gradio as gr
|
5 |
+
from groq import Groq
|
6 |
+
from deep_translator import GoogleTranslator
|
7 |
+
import pickle
|
8 |
+
from diffusers import StableDiffusionPipeline
|
9 |
+
import matplotlib.pyplot as plt
|
10 |
+
import torch
|
11 |
+
|
12 |
+
|
13 |
+
# Replace with your actual API key
|
14 |
+
api_key ="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"
|
15 |
+
client = Groq(api_key=api_key)
|
16 |
+
|
17 |
+
|
18 |
+
model_id1 = "dreamlike-art/dreamlike-diffusion-1.0"
|
19 |
+
model_id2 = "stabilityai/stable-diffusion-xl-base-1.0"
|
20 |
+
|
21 |
+
pipe = StableDiffusionPipeline.from_pretrained(model_id1, torch_dtype=torch.float16, use_safetensors=True)
|
22 |
+
pipe = pipe.to("cpu")
|
23 |
+
|
24 |
+
prompt = """dreamlikeart, a grungy woman with rainbow hair, travelling between dimensions, dynamic pose, happy, soft eyes and narrow chin,
|
25 |
+
extreme bokeh, dainty figure, long hair straight down, torn kawaii shirt and baggy jeans
|
26 |
+
"""
|
27 |
+
image = pipe(prompt).images[0]
|
28 |
+
|
29 |
+
# Function to transcribe, translate, and analyze sentiment
|
30 |
+
def process_audio(audio_path, image_option):
|
31 |
+
if audio_path is None:
|
32 |
+
return "Please upload an audio file.", None, None, None
|
33 |
+
|
34 |
+
# Step 1: Transcribe audio
|
35 |
+
try:
|
36 |
+
with open(audio_path, "rb") as file:
|
37 |
+
transcription = client.audio.transcriptions.create(
|
38 |
+
file=(os.path.basename(audio_path), file.read()),
|
39 |
+
model="whisper-large-v3",
|
40 |
+
language="ta",
|
41 |
+
response_format="verbose_json",
|
42 |
+
)
|
43 |
+
tamil_text = transcription.text
|
44 |
+
except Exception as e:
|
45 |
+
return f"An error occurred during transcription: {str(e)}", None, None, None
|
46 |
+
|
47 |
+
# Step 2: Translate Tamil to English
|
48 |
+
try:
|
49 |
+
translator = GoogleTranslator(source='ta', target='en')
|
50 |
+
translation = translator.translate(tamil_text)
|
51 |
+
except Exception as e:
|
52 |
+
return tamil_text, f"An error occurred during translation: {str(e)}", None, None
|
53 |
+
|
54 |
+
|
55 |
+
|
56 |
+
# Step 3: Generate image (if selected)
|
57 |
+
image = None
|
58 |
+
if image_option == "Generate Image":
|
59 |
+
try:
|
60 |
+
model_id1 = "dreamlike-art/dreamlike-diffusion-1.0"
|
61 |
+
pipe = StableDiffusionPipeline.from_pretrained(model_id1, torch_dtype=torch.float16, use_safetensors=True)
|
62 |
+
pipe = pipe.to("cpu")
|
63 |
+
image = pipe(translation).images[0]
|
64 |
+
except Exception as e:
|
65 |
+
return tamil_text, translation, f"An error occurred during image generation: {str(e)}"
|
66 |
+
|
67 |
+
return tamil_text, translation, image
|
68 |
+
|
69 |
+
# Create Gradio interface
|
70 |
+
with gr.Blocks(theme=gr.themes.Base()) as iface:
|
71 |
+
gr.Markdown("# Audio Transcription, Translation, and image Generate")
|
72 |
+
with gr.Row():
|
73 |
+
with gr.Column():
|
74 |
+
audio_input = gr.Audio(type="filepath", label="Upload Audio File")
|
75 |
+
image_option = gr.Dropdown(["Generate Image", "Skip Image"], label="Image Generation", value="Generate Image")
|
76 |
+
submit_button = gr.Button("Process Audio")
|
77 |
+
with gr.Column():
|
78 |
+
tamil_text_output = gr.Textbox(label="Tamil Transcription")
|
79 |
+
translation_output = gr.Textbox(label="English Translation")
|
80 |
+
image_output = gr.Image(label="Generated Image")
|
81 |
+
|
82 |
+
submit_button.click(
|
83 |
+
fn=process_audio,
|
84 |
+
inputs=[audio_input, image_option],
|
85 |
+
outputs=[tamil_text_output, translation_output, image_output]
|
86 |
+
)
|
87 |
+
|
88 |
+
# Launch the interface
|
89 |
+
iface.launch()
|