File size: 8,042 Bytes
ceda26e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ab96885
ceda26e
 
ab96885
ceda26e
 
 
 
 
 
ab96885
ceda26e
ab96885
ceda26e
ab96885
ceda26e
ab96885
ceda26e
ab96885
 
 
 
 
 
 
 
ceda26e
 
 
 
 
ab96885
ceda26e
 
 
 
 
ab96885
ceda26e
ab96885
ceda26e
ab96885
ceda26e
ab96885
ceda26e
ab96885
 
 
 
 
 
 
 
 
ceda26e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
import gradio as gr
from transformers import T5Tokenizer, T5ForConditionalGeneration, BartTokenizer, BartForConditionalGeneration
import requests
from bs4 import BeautifulSoup
import gtts
from io import BytesIO
import base64
import os

# Function to fetch text from a URL
def fetch_text_from_url(url):
    try:
        response = requests.get(url)
        soup = BeautifulSoup(response.content, 'html.parser')
        paragraphs = soup.find_all('p')
        text = ' '.join([para.get_text() for para in paragraphs])
        return text, None
    except Exception as e:
        return None, f"Error fetching URL: {e}"

# Function to summarize text using T5
# Function to summarize text using T5
def summarize_t5(text, size):
    model_name = "C:\\Users\\zurin\\Desktop\\text summarization\\fine_tuned_t52"
    tokenizer = T5Tokenizer.from_pretrained(model_name)
    model = T5ForConditionalGeneration.from_pretrained(model_name)
    
    input_text = f"summarize: {text}"
    inputs = tokenizer(input_text, return_tensors="pt", max_length=512, truncation=True)
    
    # Define length parameters
    if size == "Short":
        min_len, max_len = 30, 50
    elif size == "Medium":
        min_len, max_len = 50, 100
    else:  # Long
        min_len, max_len = 100, 200
        
    summary_ids = model.generate(
        inputs["input_ids"],
        max_length=max_len,
        min_length=min_len,  # Use the specified min_length instead of fixed 10
        length_penalty=1.0,  # Reduced from 2.0 to allow more length variation
        num_beams=4,
        early_stopping=True
    )
    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
    return summary

# Function to summarize text using BART
def summarize_bart(text, size):
    model_name = "C:\\Users\\zurin\\Desktop\\text summarization\\fine_tuned_bart"
    tokenizer = BartTokenizer.from_pretrained(model_name)
    model = BartForConditionalGeneration.from_pretrained(model_name)
    
    inputs = tokenizer(text, return_tensors="pt", max_length=1024, truncation=True)
    
    # Define length parameters
    if size == "Short":
        min_len, max_len = 30, 50
    elif size == "Medium":
        min_len, max_len = 50, 100
    else:  # Long
        min_len, max_len = 100, 200
        
    summary_ids = model.generate(
        inputs["input_ids"],
        max_length=max_len,
        min_length=min_len,
        length_penalty=0.8,  # Reduced from 1.0 to encourage length variation
        num_beams=6,
        no_repeat_ngram_size=2,  # Added to prevent repetition
        early_stopping=True
    )
    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
    return summary
# Function to convert text to speech and save as a file
def text_to_speech(text):
    tts = gtts.gTTS(text)
    audio_file_path = "summary_audio.mp3"
    tts.save(audio_file_path)
    return audio_file_path

# Main function to handle summarization
def summarize_news(input_type, text_input, url_input, model_choice, size_choice):
    # Determine the input text based on the input type
    if input_type == "Text":
        if not text_input:
            return "Please provide text to summarize.", None, None
        input_text = text_input
    else:  # URL
        if not url_input:
            return "Please provide a URL to summarize.", None, None
        input_text, error = fetch_text_from_url(url_input)
        if error:
            return error, None, None

    # Summarize the text
    if model_choice == "T5":
        summary = summarize_t5(input_text, size_choice)
    else:  # BART
        summary = summarize_bart(input_text, size_choice)

    # Generate audio for the summary
    audio_file = text_to_speech(summary)

    return summary, audio_file, None

# Custom CSS for the design
custom_css = """
<style>
/* Background for the entire app */
body {
    background: linear-gradient(135deg, #E6E6FA 0%, #D8BFD8 100%) !important;
    font-family: 'Arial', sans-serif;
    min-height: 100vh;
    margin: 0;
    display: flex;
    justify-content: center;
    align-items: center;
}
/* White container for all elements */
.container {
    background-color: #FFFFFF !important;
    border-radius: 15px !important;
    padding: 30px !important;
    margin: 20px auto !important;
    max-width: 800px !important;
    box-shadow: 0 4px 10px rgba(0, 0, 0, 0.1) !important;
    width: 100%;
}
/* Title styling */
.title {
    font-size: 36px;
    color: #000000 !important;
    text-align: center;
    font-weight: bold;
    margin-bottom: 10px;
}
/* Subtitle styling */
.subtitle {
    font-size: 18px;
    color: #000000 !important;
    text-align: center;
    margin-bottom: 20px;
}
/* Labels for inputs */
label {
    color: #000000 !important;
}
/* Input and textarea styling */
input, textarea {
    background-color: #F5F5F5 !important;
    color: #000000 !important;
    border: 1px solid #D3D3D3 !important;
    border-radius: 5px !important;
}
/* Dropdown styling */
select {
    background-color: #F5F5F5 !important;
    color: #000000 !important;
    border: 1px solid #D3D3D3 !important;
    border-radius: 5px !important;
    padding: 5px !important;
}
/* Button styling */
button {
    background-color: #9370DB !important;
    color: white !important;
    border-radius: 10px !important;
    padding: 8px 20px !important;
    border: none !important;
    display: block !important;
    margin: 20px auto !important;
    cursor: pointer !important;
}
button:hover {
    background-color: #4B0082 !important;
}
/* Footer styling */
.footer {
    text-align: center;
    color: #000000 !important;
    font-size: 14px;
    margin-top: 30px;
}
.footer-heart {
    color: #FF0000 !important;
}
/* Output text and error messages */
.output-text, .error-text {
    color: #000000 !important;
}
</style>
"""

# Gradio app
with gr.Blocks() as app:
    # Inject custom CSS
    gr.HTML(custom_css)

    # Main container
    with gr.Column(elem_classes=["container"]):
        # Title and subtitle
        gr.HTML('<p class="title">BBC News Summarizer</p>')
        gr.HTML('<p class="subtitle">Summarize news articles with T5 or BART in your preferred length!</p>')

        # Input section
        input_type = gr.Radio(choices=["Text", "URL"], label="Choose input type:", value="Text")
        
        with gr.Row():
            text_input = gr.Textbox(label="Enter news text here:", lines=5, visible=True, placeholder="Paste your news text here...")
            url_input = gr.Textbox(label="Enter news URL here:", visible=False, placeholder="Enter a news article URL...")

        # Show/hide text input or URL input based on input type
        def update_input_visibility(input_type):
            return (
                gr.update(visible=(input_type == "Text")),
                gr.update(visible=(input_type == "URL"))
            )

        input_type.change(
            fn=update_input_visibility,
            inputs=input_type,
            outputs=[text_input, url_input]
        )

        # Model selection
        model_choice = gr.Dropdown(choices=["T5", "BART"], label="Choose summarization model:", value="T5")

        # Summary size selection
        size_choice = gr.Dropdown(choices=["Short", "Medium", "Long"], label="Choose summary size:", value="Short")

        # Summarize button
        summarize_button = gr.Button("Get Summary")

        # Outputs
        summary_output = gr.Textbox(label="Summary:", elem_classes=["output-text"])
        audio_output = gr.Audio(label="Listen to the Summary:")
        error_output = gr.Textbox(label="Error:", elem_classes=["error-text"], visible=False)

        # Footer
        gr.HTML('<p class="footer">Powered by xAI\'s Grok | Made with <span class="footer-heart">❤️</span> for news enthusiasts</p>')

        # Button click event
        summarize_button.click(
            fn=summarize_news,
            inputs=[input_type, text_input, url_input, model_choice, size_choice],
            outputs=[summary_output, audio_output, error_output]
        )

# Launch the app
app.launch()