siddhartharya
commited on
Commit
•
416840c
1
Parent(s):
e176e7d
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import pipeline
|
3 |
+
import PyPDF2
|
4 |
+
import docx
|
5 |
+
import requests
|
6 |
+
|
7 |
+
# Load the Llama model from Groq Cloud
|
8 |
+
model = pipeline("text-to-speech", model="groq/llama-3.1-70b")
|
9 |
+
|
10 |
+
def extract_text_from_file(file_path):
|
11 |
+
if file_path.endswith('.pdf'):
|
12 |
+
with open(file_path, 'rb') as file:
|
13 |
+
reader = PyPDF2.PdfReader(file)
|
14 |
+
text = ""
|
15 |
+
for page in reader.pages:
|
16 |
+
text += page.extract_text()
|
17 |
+
return text
|
18 |
+
elif file_path.endswith('.docx'):
|
19 |
+
doc = docx.Document(file_path)
|
20 |
+
text = "\n".join([para.text for para in doc.paragraphs])
|
21 |
+
return text
|
22 |
+
return ""
|
23 |
+
|
24 |
+
def fetch_text_from_url(url):
|
25 |
+
response = requests.get(url)
|
26 |
+
return response.text if response.status_code == 200 else ""
|
27 |
+
|
28 |
+
def generate_podcast(uploaded_file=None, url=None):
|
29 |
+
if uploaded_file:
|
30 |
+
input_text = extract_text_from_file(uploaded_file.name)
|
31 |
+
elif url:
|
32 |
+
input_text = fetch_text_from_url(url)
|
33 |
+
else:
|
34 |
+
return "No input provided."
|
35 |
+
|
36 |
+
dialogue = f"Man: {input_text}\nWoman: {input_text} (response)"
|
37 |
+
audio = model(dialogue)
|
38 |
+
return audio
|
39 |
+
|
40 |
+
iface = gr.Interface(
|
41 |
+
fn=generate_podcast,
|
42 |
+
inputs=[
|
43 |
+
gr.inputs.File(label="Upload PDF or DOC"),
|
44 |
+
gr.inputs.Textbox(label="Or enter URL"),
|
45 |
+
],
|
46 |
+
outputs=gr.outputs.Audio(label="Generated Podcast"),
|
47 |
+
)
|
48 |
+
|
49 |
+
iface.launch()
|