File size: 7,808 Bytes
d3319a1 abc4408 d3319a1 abc4408 1cd73aa abc4408 43c84e4 abc4408 d3319a1 abc4408 91b29f9 440327e 91b29f9 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 d3319a1 abc4408 |
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 |
import streamlit as st
import validators
import os
from dotenv import load_dotenv
from langchain_groq import ChatGroq
from langchain_core.prompts import PromptTemplate,ChatPromptTemplate,MessagesPlaceholder
from langchain.chains.summarize import load_summarize_chain
from langchain_community.document_loaders import YoutubeLoader,UnstructuredURLLoader
from PIL import Image
from langchain_core.output_parsers import StrOutputParser
from pytube.innertube import InnerTube
from langchain_community.tools import DuckDuckGoSearchRun
from langchain_community.utilities import DuckDuckGoSearchAPIWrapper
from langchain.agents import create_openai_tools_agent,AgentExecutor
load_dotenv()
# api keys
groq_api_key = os.getenv('GROQ_API_KEY')
# tool
api_wrapper = DuckDuckGoSearchAPIWrapper(max_results=3,safesearch='none')
tool = DuckDuckGoSearchRun(api_wrapper=api_wrapper)
# app
img1 = Image.open('youtubeLogo.jpg')
img2 = Image.open('web.jpeg')
st.set_page_config(page_title='Summarizer')
col1,col2,col3 = st.columns([1,1,3])
with col1:
st.image(img1,use_column_width=True)
with col2:
st.image(img2,use_column_width=True)
with col3:
st.title('Summarize Youtube Videos / Web Page')
llm = ChatGroq(model='Gemma2-9b-It',groq_api_key=groq_api_key)
system_prompt = '''
Summarize the content from given {text} in best possible way.
'''
prompt = PromptTemplate(
input_variables=['text'],
template=system_prompt
)
system_prompt_n = '''
You are an advanced AI summarization tool .Summarize the {text} in the best possible way.
Analyze the content, focusing on the following criteria:
1.Main Ideas: Identify the primary arguments or messages conveyed in the content.
2.Supporting Details: Highlight important examples, statistics, or pieces of evidence that reinforce the main ideas.
3.Structure: Maintain the original flow of the content, summarizing it in a coherent manner that mirrors the order of presented information.
4.Length: Produce a summary that is concise.
5.Clarity and Readability: Ensure the summary is written in clear, simple language that is easy to understand, avoiding difficult words unless necessary.
6.Contextual Relevance: Tailor the summary based on the intended audience, making sure it is appropriate for their level of understanding and interest.
'''
final_prompt = PromptTemplate(
input_variables=['text'],
template=system_prompt_n
)
lyrics_prompt = '''
You're a skilled data retrieval assistant with expertise in extracting and processing information from various online sources,
especially youtube videos. Your focus is on accurately identifying and presenting song lyrics as well as the
artist of the song,singer of the song ,the lable company and the song writer from {text},
ensuring clarity and precision in the output.Your task is to fetch the lyrics and singer,artist,label and song writer from the {text}
generated from a YouTube link of a music video.Don't provide lyrics in paragraph form, provide line by line with proper spacing.
Keep in mind that you should provide the lyrics in the {language} without any additional commentary or unrelated information.
'''
lyrics_template = PromptTemplate(
input_variables=['text','language'],
template=lyrics_prompt
)
tool_sys_prompt = '''
you are helpful assistant capable of providing full lyrics in poem format as well as singer,artist,label
and writer of the {song} in {language}.
'''
# Tools and agent
tool_template = ChatPromptTemplate([
('system',tool_sys_prompt),
MessagesPlaceholder('chat_history',optional=True),
('user','{song}'),
('user','{language}'),
MessagesPlaceholder('agent_scratchpad')
])
agent = create_openai_tools_agent(llm=llm,tools=[tool],prompt=tool_template)
executor = AgentExecutor(agent=agent,tools=[tool],verbose=False)
# url loader and text extractor function
def loadAndExtract():
if 'youtube.com' in URL or 'youtu.be' in URL:
loader = YoutubeLoader.from_youtube_url(URL,add_video_info=True,language=['en-IN','hi','en'],)
else:
loader = UnstructuredURLLoader([URL],ssl_verify=False,mode='single',
headers={'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36 BingBot/2.0'})
txt = loader.load()
return txt
# main app content
URL = st.text_input(label=':blue[**Paste URL here**]',placeholder='https://www.youtube.com/')
option = [':orange[**Select**]',':orange[**English**]',':orange[**Hindi**]']
if "button_pressed" not in st.session_state:
st.session_state.button_pressed = None
if st.sidebar.button('Previous'):
st.session_state.button_pressed = None # Reset state when pressing 'Previous'
st.rerun()
if st.session_state.button_pressed != 2:
if st.button('Summarize the Video/Web-page'):
st.session_state.button_pressed = 1
if st.session_state.button_pressed != 1:
if st.button('Provide Lyrics of the song'):
st.session_state.button_pressed = 2
selected_option = st.radio(label=':blue[**Select any one language for lyrics**]',options=option)
if st.session_state.button_pressed == 1:
if not URL:
st.info('Please provide the url')
elif not validators.url(URL):
st.warning('Please provide the correct url')
else:
try:
with st.spinner(':pink[summarizing]'):
txt = loadAndExtract()
chain = load_summarize_chain(llm,chain_type='map_reduce',
map_prompt=prompt,
combine_prompt=final_prompt)
response = chain.run(txt)
st.success(response)
except Exception as e:
st.exception(f'Exception : {e}')
elif st.session_state.button_pressed == 2:
if not URL:
st.info('Please provide the url')
elif not validators.url(URL):
st.warning('Please provide the correct url')
else:
try:
if 'youtube.com' in URL or 'youtu.be' in URL:
loader = YoutubeLoader.from_youtube_url(URL,add_video_info=True,language=['en-IN','hi','en'],)
txt = loader.load()
if txt == []:
user_input = st.text_input(label='Plz provide name of the song')
if user_input:
with st.spinner(':pink[lyrics on the way]'):
if option.index(selected_option) == 1:
language = 'english'
elif option.index(selected_option) == 2:
language = 'hindi'
res = executor.invoke({'song':user_input,'language':language})
st.success(res)
else:
chain = lyrics_template|llm|StrOutputParser()
with st.spinner(':pink[lyrics on the way]'):
if option.index(selected_option) == 1:
language = 'english'
res = chain.invoke({'text':txt,'language':language})
st.success(res)
elif option.index(selected_option) == 2:
language = 'hindi'
res = chain.invoke({'text':txt,'language':language})
st.success(res)
except Exception as e:
st.exception(f'Exception {e}')
|