Discover the innovative AI applicationss that have been built during lablab.ai hackathons. Explore our AI Hackathon projects and get inspired to create your own!
This app is a powerful tool designed to assist content creators who have a substantial following of over 5000 viewers or followers. By leveraging the YouTube API, the app can efficiently read through all the comments and classify them according to their sentiment, toxic content, or spam. After classifying the comments, the app creates a word embedding using cohere AI, which helps to identify patterns and group similar comments together. These comments are then grouped into three clusters using KMeans. By clustering the comments, content creators can quickly analyze and understand the nature of the comments and identify any common themes or topics that resonate with their audience. Using the analysis of the clustered comments, the app generates a summary of the main themes and extracts the primary topics. These topics are presented to the content creators, along with suggested content ideas for future posts. This helps content creators to understand the interests and preferences of their audience better and create content that resonates with their viewers. In conclusion, this app provides content creators with a streamlined approach to understanding their audience's preferences and interests by analyzing the comments on their posts. By classifying, clustering, and summarizing comments, the app provides a valuable tool for generating new content ideas that will engage and excite their audience, leading to increased views and followers.
For non-native English speakers conducting business with foreign companies or entities, there are several pain points that can arise in the email communication process. One of the major pain points is the fear of using incorrect grammar or inappropriate language, which can cause confusion or even offend the recipient. Non-native English speakers often struggle with finding the right words and phrases to express themselves effectively in English, which can lead to misunderstandings and miscommunications. Another pain point is the time-consuming process of generating unique content for the same clients and topics. Many non-native English speakers find themselves repeating the same information or ideas in different emails, which can be both tedious and unprofessional. Lastly, conveying ideas effectively can be a significant challenge, especially in bi-lingual situations where cultural and linguistic differences can create additional barriers. Finding the right balance between professionalism and appropriate tone can be difficult, and non-native English speakers may struggle to communicate their ideas in a way that is both clear and culturally sensitive. Mars Mail addresses these pain points by providing an AI-powered email assistant that helps non-native English speakers in East Asia quickly and efficiently produce high-quality, context-appropriate, and polite responses for their business emails. By using a fine-tuned GPT3.5 model, Mars Mail is specifically designed to handle bi-lingual business email situations with high accuracy, reducing the fear of incorrect grammar and language usage. The user-friendly chatbot interface also makes it easy for non-technical users to communicate their ideas effectively, saving time and improving the quality of email communication.
We believe that AI can help SMBs be more successful by saving time, generating, and reducing hiring cost. SMBs find social media challenging, time-consuming, and the alternatives to doing it by themselves are hiring a social media manager or ads agency which are costly. This is exacerbated by complex tech they need to navigate in doing this, such as understanding social media algorithms and data analyses on what performs better. We interviewed a pottery artist who said "Making Instagram Reels sucked the joy out of creating." We envision an AI social media manager that is affordable, effective and smart, while also always reliable, loyal and active. The AI social media manager will create more engaging and sales-generating contents in seconds instead of hours. All this can be done using text / natural language interface and data optimization in the background so SMBs do not have to navigate complex technical details. We also plan to improve the AI to the point that it can run SMBs social media independently, just like if you were to hire a coworker. This will enable SMBs to enjoy what they are doing more, be more efficient when doing business, and save business owners from stressful mistakes or failures.
Our goal is to develop a system that can take documents in any language and allow users from different parts of the world with different languages to interact with those documents. This can help spread knowledge without any language barrier. We are first testing it out with Quranic text, which is in Arabic, and building a chatGPT-style bot for question answering. In future, we would like to expand it to additional documents in any language. We first perform embedding of the text using Cohere's multilingual embed model and save those embeddings in a vector database, Pinecone DB in our case. We then take user queries and also embed them and perform a similarity-based search to provide the most relevant results based on the query. The app is currently deployed and publicly available. Streamlit app link: https://taqihaider7-tafsir-quran-sementic-search-llm-app-fwh2if.streamlit.app/
The existing search bar in most websites performs keyword search. A slow and arduous process in which the user has to read through a myriad of information before finding the tidbit that they wanted in the first place. My goal was to create a question answering tool that can be easily integrated into any website. It allows you to find specific information, provides answers in a clear, understandable way and includes sources and more information should the user need it. Presenting Web Indexer. Developed to significantly improve the user experience by providing a service that ChatGPT, Google and standard search bars cannot. This project was made for the AI21 labs hackathon organized by lablab.ai.
Our solution is a smart Slack bot that implements multilingual semantic search and sentiment analysis to facilitate multi-language messaging platforms by improving the chat search function. The bot allows users to search using any language for messages in any language. It also contains a sentiment analysis that gives moderators a sentiment report on the messages of specific users or the chat as a whole.
Our application aims to make it easier for users to find and access the information they need within the specified video. By using the Cohere API, we are able to perform advanced semantic searches that go beyond simple keyword matching, allowing users to find content relevant to their query in the specified video. Our application utilizes the Cohere API to locate and highlight the specific transcript within the video that matches their search. This makes it easier for users to quickly find and access the information they need, without having to manually search through the entire video. Overall, our application is a game-changer for anyone looking to find and access specific information within YouTube videos. Whether you're a student looking for a specific lecture, a researcher trying to locate a specific quote, or just someone looking for information on a particular topic, our application makes it easier than ever to find what you need.
People with dyslexia often find it hard to read and write, primarily because it’s “hard for them to mentally lock in” as described by the person suffering from Dyslexia. This makes studying a struggle for them however it is seen that in most cases people with this condition find it easier to read on electronic devices rather than the real document itself. Although it’s helpful to have a guide to help out when difficulties arise. But what if a person doesn’t have access to a guide? This is where NotAlone comes in to help. Dyslexia can be an obstacle but with LLMs and Deep Learning making people’s life easier let’s take a step to make it more accessible to everyone so that barriers like these don’t dominate a person’s will to learn and write. NotAlone is specifically designed to empower individuals with dyslexia by providing a seamless learning-rich writing environment tailored to their unique needs. Our goal is to ensure that no one feels left behind in today's fast-paced world. People with dyslexia often prefer speaking over writing, hence many take the help of an ASR app to help them do so. Inspired by this, we provide a Whisper-based STT feature to help them type by speaking. ChatGPT-based writing assistance is another essential feature of NotAlone. This feature provides personalized guidance and support by helping users overcome challenges in writing and reading by:- 1. Helping them write about anything they want 2. Grammar Correction 3. Rewriting 4. Explaining a word/phrase 5. Summarize a paragraph 6. Suggest Synonyms 7. Chat-based assistance Even though we help them read better but some words are complicated even for us to understand this is where Text-to-Speech service can help them read a word or a paragraph whenever they feel stuck. Beyond these core features, we provide an interface that allows users to adjust settings, such as line height, word spacing, and background color. All this along with custom fonts that cater specifically to dyslexic people.
At Dripper News, we know that tech industry CEOs are always on the go. That's why we've created a mobile-focused news aggregator that makes staying informed easier than ever. Our platform curates and filters the latest and most relevant tech news from reputable sources, using cutting-edge AI technology to save you time and effort. But that's not all - our platform also creates a personalized news feed and podcast tailored to your interests, so you can stay ahead of the competition with the latest trends, industry developments, and competitor updates. With Dripper News, you can make informed decisions for your company and stay on top of the ever-changing tech industry, all in one convenient place.
Intelligent Health Assistant is a groundbreaking app that utilizes AI technology to help patients with their symptoms before they see a doctor. The app records symptoms in a 10-second timeframe and transcribes them using Whisper API to store them in a file. The app then uses ChatGPT API to check the history of previous inputs, outputs, and current input, and then analyzes the symptoms and compares them to a vast database of medical information to identify potential illnesses and conditions. The app is designed to guide people who do not prioritize going to a doctor or who feel worried about their symptoms, and advise them on the urgency of their symptoms. This app can help reduce the number of patients who ignore medical symptoms, as well as help identify potential illnesses and conditions in their early stages.
Deaf and hard-of-hearing individuals face a multitude of challenges in their daily lives, with one of the biggest being the difficulty in communicating with hearing individuals who do not understand sign language. This communication barrier can lead to social isolation, limited access to education and employment opportunities, and a lack of participation in various social activities. Our project aims to address this challenge by developing a real-time speech-to-sign language translation solution that can bridge the communication gap between the deaf and hard-of-hearing and the hearing individuals. This solution has the potential to enhance accessibility and inclusivity for the deaf and hard-of-hearing community and improve their quality of life.
1.It can deliver near human like conversational experience. 2.It can solve easy level coding questions. 3.Available for 24/7. 4.It is a computer program that uses artificial intelligence to conduct a conversation with users, typically via a messaging interface. Chatbots can be used for a variety of purposes, such as customer service, sales, or information retrieval. 5.It can be integrated with various messaging platforms, such as Facebook Messenger, WhatsApp, or Slack, and can also be embedded on websites or mobile apps. They can be customized to suit specific business needs and can help automate routine tasks, reduce response times, and improve customer satisfaction.
The problem I want to solve with this program is to increase my efficiency when studying for exams. Often I feel overwhelmed with my different forms of study material like lecture recordings, notes, slides, pdf's, or voice recordings. With this program I am attempting to import all of my material in one place and then, with the help of AI, create summaries based on my content. I can then enter a interactive world where my program creates tailored exercises for me in order to prepare for my exam. Perhaps I can even feed my program with a past exam at one point and it then generates another exam in the style of the previous exam. I believe this project has great potential and many use cases for students, but also for any individual that is trying to test their knowledge on their chosen topic.
Generate a rap song for any YouTube video Music is easier for people to accept information than articles. If there is a news or introduction video on YouTube, we only need to enter the YouTube link, and a rap song will be automatically generated. The process is as follows: Enter the YouTube URL Enter the background music of the YouTube instrumentals Enter the rap lyrics style First, the video will be converted into text using the Whisper API, and then the GPT3.5 API will condense and highlight the key points of the text. These key points will be written into lyrics in a certain style. Next, we use Python to download the background music of the YouTube instrumentals, preprocess the lyrics, capture the rhythm of the music, match the words with the heavy beats, and then use gTTS to speak the words and automatically adjust the audio position. Then your rap song will be generated!
Clients go to our website, where they can paste the URL or video itself and select the desired language for translation. The model using GPT-3 automatically determines the language for generation, translates into the required language. Also, the model automatically determines intonation, pronunciation speed, age, gender of the speaker and could make sample of own voice. The client receives the video in the language selected at the beginning. For the future, we want to link our site to virtual assistants so that videos are accessible to people with disabilities. And also connect the ability to translate into all languages of the world
This project aims to revolutionise the quality & assurance auditing process of call centres by transcribing and analysing audio recordings using GPT-3 and Whisper models. Call centres handle an immense volume of customer interactions everyday, and it's crucial for businesses to evaluate the quality of these interactions to maintain high customer satisfaction rate. Traditionally, quality and assurance auditing has been a time-consuming and manual process, where human auditors listen to and evaluate customer calls. However, this approach is prone to human error, inconsistency, and scalability challenges. The proposed solution leverages Automated Speech Recognition (ASR) and Large Language Model to automate and streamline the quality and assurance auditing process. The system summarises key information in calls, such as issues and solutions, and other relevant data points. Additionally, the solution conducts sentiment analysis to evaluate the tone and mood of the conversation, as well as evaluate customer experience and satisfaction levels. In addition, it provides feedback and insights into the performance of the call centre staff and identify areas for improvement. Overall, the proposed solution has the potential to transform the call centre industry, providing businesses with valuable insights into their customer interactions and enabling them to take proactive steps to improve their overall customer experience.
Polly was made possible using GPT 3.5, Whisper, and Google's Text-to-Speech. Putting these components together enables us to communicate in many forms of media and practice in different ways. The current implementation uses Whisper and Google's Text-to-Speech pair to receive and output voice messages to enhance interactive learning. We implemented six languages to choose from as we experimented with how each implementation is performed. There are many improvements within this field; this is a new way to practice and brush up your skills in Deutsche, French, English, etc. Polly is a project we intend to pursue further and in multiple different use cases.
AURA-the chatbot is a chatgpt-integrated speech assistant that converts user audio input into text and feeds it to chatgpt.Chatgpt then discovers a solution to our query which is converted back into audio files by Auro for the users. The two main tools used for this project are whisper and chatgpt API. Whisper can accurately recognize speech in a multitude of languages, accents, and environments. It can handle technical language and background noise and perform at par with human capabilities. While ChatGPT is a specialized variant of the GPT-3 language model designed to generate human-like responses in conversational contexts. We combine both of these cutting-edge technologies to develop an API that is useful to everyone.
Introducing our innovative Streamlit application, which harnesses the power of OpenAI GPT-3 to generate multi-layer encryption and decryption codes for secure communication. This application is designed to help users easily encrypt and decrypt their messages using state-of-the-art encryption techniques, making it nearly impossible for unauthorized parties to access their sensitive information. To use this application, users can input their speech message through OpenAI Whisper, which transcribes the message accurately. The application then uses GPT-3 to generate a multi-layer encryption code, which can be customized by the user according to their specific requirements. Once the encryption code is generated, it is applied to the speech message, making it indecipherable to anyone without the decryption code. Users can choose from a variety of encryption algorithms and key lengths, and can also input their own unique encryption key for added security. The application also allows users to save and retrieve their encryption codes for future use, making it easy to communicate securely with their contacts. In addition to its powerful encryption capabilities, the application is also highly user-friendly, with a clean and intuitive interface that allows users to easily navigate and customize their encryption settings. With its cutting-edge technology and ease of use, this Streamlit application is the perfect solution for anyone looking to communicate securely and confidently in today's digital world.
Our application generates full courses on various topics and provides video lectures with text explanations and generated questions to reinforce learning. With the help of our AI mentor, students can learn at their own pace and receive personalized feedback to improve their understanding. One of the most exciting features of our application is the simple gamification process that engages students in the learning process. By providing a fun and interactive learning environment, we believe that students will be more motivated to learn and retain the information for longer periods. We are also proud to announce that our application utilizes Synthesia AI to generate automated videos based on the GPT-3 script. This ensures that our video lectures are of the highest quality and can be easily accessed by students on any device. Another major benefit of using AI to generate our courses is that each instance of the course is unique and up-to-date with the latest features and technologies. Unlike traditional courses that may become outdated over time, our AI mentorship application ensures that students are learning the most relevant and current information. With the help of OpenAI GPT-4, our application can quickly and efficiently generate new course material as new technologies and features emerge. This means that our students will always be on the cutting edge of their respective fields, giving them a competitive advantage in the job market.
The Multilingual Book-based Information Retrieval System is an AI-powered tool that allows users to extract relevant information from a given book in any language. The system works by first identifying the user's input keywords or sentences related to a specific topic. It then uses natural language processing and machine learning algorithms to search the relevant book for information on the given topic. The system also provides a translation feature that allows the user to translate the retrieved information and references into any language of their choice. This feature makes it easier for users who are not familiar with the language of the book to access the information they need. The Multilingual Book-based Information Retrieval System is beneficial for researchers, students, and anyone who needs to access information from books in different languages. With this system, users can easily retrieve relevant information from books without having to spend hours searching through pages. The system is designed to provide accurate and reliable information in a user-friendly interface, making it a valuable tool for anyone seeking knowledge on different topics.
Maverick REACT offers artificial intelligence integration for emergency situations. Our service uses AI with the necessary event information provided by government officials and acts as an assistant to provide key protocols and information to citizens. The AI service is accessed via SMS or web portal, offering a solution without internet. How does our service work? When an emergency situation occurs, such as a flood, fire or earthquake, our service sends an SMS message or makes a voice call to numbers registered in a database or the citizen can contact a number provided by the authorities. The message or call contains information about the type and severity of the emergency, preventive measures that should be taken and resources available in the area. The user can respond to the message or call with specific questions about their personal situation or request additional help. Our service uses AI algorithms to process responses and offer personalized and updated advice. REACT has several advantages over traditional emergency alert and response systems. Firstly, it does not depend on the internet, which means it can function even when there are power outages or problems with mobile networks. Secondly, REACT service is interactive and adaptable to the individual needs of each user. Thirdly, it uses reliable and verified sources of information provided by the government or other authorized organizations. And finally REACT is fast and efficient in sending and receiving large-scale messages or calls. Our goal is to contribute to creating a safer and more resilient world in the face of emergency situations through innovative and intelligent use of technology. We believe that our service can save lives and reduce suffering caused by disasters. If you want to know more about our service or how to register for it, contact us. We are Maverick AI.
In a post truth world, the ability to quickly fact check any information against the available official public records and statements is a superpower that can hold public figures accountable, make gaslighting the public during election time more difficult and save democracies. It also cuts down distractions and time required to do research on public figures and reduces the entry barrier for political journalism and fact checking. Twitter is the internet’s town house and most of the public figures use it so the ability to fact check them in real time is critical. We aim to empower the general public with tools that make the world a better place by keeping people in power accountable.
a personalized news feed focused on the tech industry, powered by artificial intelligence (AI). Our news aggregator is specifically designed for busy CEOs, providing them with the latest and most relevant news in the tech sector. Through the use of AI, our platform curates and filters news articles from reputable sources, presenting only the most important and timely news stories to our users. This allows CEOs to stay informed on the latest trends, industry developments, and competitor updates in a quick and efficient manner. Additionally, our news aggregator provides a personalized experience for each user. By analyzing the user's reading habits and interests, our AI technology tailors the news feed to provide a custom selection of articles that are most relevant to their business and industry. Overall, our personalized news feed offers a comprehensive solution for CEOs who want to stay informed on the latest developments in the tech industry without the hassle of sorting through countless news sources. With our platform, CEOs can stay ahead of the curve and make informed decisions for their companies.
Legal-ease is a web-app hosting a suite of tools useful for simplifying legal documents for common use. In this way, Legal-ease offers tools to perform: 1. QnA over legal documents 2. Document summarization 3. Cross-lingual search This is great for people who dont have the time, are not native speakers of english or otherwise struggle with legal documents on a day to day basis. It is also useful for lawyers and researchers with access to a large collection of legal documents. Summarization is handled via cohere, QnA is handled via langchain, cohere's embeddings and LLM, and qdrant's excellent vector DB search.
Scoper is a multilingual semantic AI tool for searching across internal documents. Although you may be able to search for exact keywords, we require an intelligent embedding of strings (in this case we use Cohere’s endpoint) to have similarity based on meaning, not just text. For example, the phrase “save cost by switching” may not show up directly in any internal documents, but there are mentions in several documents about benefits of switching programming languages or models of work.Additional features are documentation classification, where we label all internal documents both as the type of document (wiki, project proposal, or memo) but also with their completion status, i.e. finished or unfinished. We also support document generation, since in order to keep moving a company quickly, having AI help draft proposals or wiki’s goes a long way.
Currently, the most popular corporate knowledge management system is Confluence by Alatasian. It is known for a lack of search capabilities and makes most corporate knowledge inaccessible, especially in fast-growing companies where regular structure and responsibilities change. Some independent vendors fill this gap by offering carefully tuned solar-based search engines for Confluence, but not real semantic search. Confluence is a proprietary cloud-based solution, and it would be difficult to MVP a search extension in a hackathon. The most advanced open-source alternative is wiki.js, which already supports external search engines. So the current goal is to implement an external search engine for wiki.js using Cohere's LLM-powered Multilingual Text Understanding model and Qdrant's vector search engine. At the second stage of the project (most likely outside the hackathon scope), we plan to add the capability to upload and index videos in our knowledge management system. Recordings of presentations and meetings are the richest source of knowledge, but they were left outside knowledge management due to technical difficulties. Simple transcription and semantic search of that content could significantly boost corporate knowledge accessibility.