🗓️ 48 hours of building (24-26th March) 💻 Create innovative new apps with OpenAI's latest ChatGPT and Whisper APIs 💡 Learn from top AI professionals ⚒️ Combine new GPT 3,5 Turbo and Whisper APIs to build your AI app 🐱💻 Now is the time to register and let's get started!
Our AI hackathon brought together a diverse group of participants, who collaborated to develop a variety of impressive projects based on:
2106
Participants
178
Teams
31
AI Applications
ChattyRental is a revolutionary chatbot powered by AI, designed to simplify and enhance the room rental experience for both renters and rental companies. Our chatbot is equipped with advanced natural language processing capabilities, enabling users to book a room through a simple, conversational interface. Additionally, ChattyRental generates personalized recommendations for rental rooms based on a user's preferences and past behavior. With our intelligent search feature, users can find the right room that meets their needs quickly and easily. By leveraging the power of ChattyRental, room rental companies can cut their sales department costs by an average of 36%, while simultaneously improving customer satisfaction and user experience.
Room Booking AI Assistant
Investing in the right talent is crucial to the success of any company, but finding candidates who are skilled and fit well with the organization's culture can be daunting. We have the solution. Our cultural fit bot, Cultivate, leverages the power of AI to help companies identify candidates who align with their culture, work style, and values. By automating the initial interview process, Cultivate enables recruiters to evaluate more candidates and make data-driven hiring decisions without compromising the quality of the hire. With Cultivate, recruiters and hiring managers can analyze candidates' responses to real-world scenarios and identify critical thinking, problem-solving, and communication skills. The bot also provides instant feedback on candidates' cultural fit, including their attitudes, beliefs, and behavioral patterns. It enables recruiters to identify top candidates that align with their company culture and values. Our platform continues beyond there. Cultivate also provides: An automatic interview summary for each candidate. Highlighting their strengths and weaknesses. A set of metrics that help recruiters make informed hiring decisions based on data, not opinions. These features significantly reduce the risk of bias in the hiring process, promoting fairness and inclusion. Investing in Cultivate means investing in the future of your company. By leveraging the power of AI to streamline the hiring process, we help you find top talent that not only meets the job requirements but also aligns with your company culture and values, ultimately improving productivity, employee engagement, and retention rates. Join us today and start growing your team with harmony and success.
Cultivate
Call centers handle an immense volume of customer interactions every day, 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. This approach is prone to human error, inconsistency, and scalability challenges. The Voice Analytics with AI aims to revolutionize the quality & assurance auditing process of call centers by transcribing and analyzing audio recordings using GPT-3 and Whisper models. The proposed solution leverages Automated Speech Recognition (ASR) and Large Language Models (LLM) to automate and streamline the quality and assurance auditing process. First, the system summarizes key information in call recordings, such as operator's name, issues, and solutions, and other relevant data points. After that, the solution conducts sentiment analysis to evaluate the tone and mood of the conversation using NLP and LLM. In addition, LLM evaluates customer experience and satisfaction levels and provides scores for each. Last but not least, the model ends the report of each call with feedback and insights about the performance of operator and suggests areas for improvement. Overall, the proposed solution has the potential to transform the call center industry, providing businesses with valuable accurate insights into their customer interactions and enabling them to take proactive steps to train their operators and improve their overall customer experience.
AI Alliance
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.
Mentorsko
Pawel Czech
Co-founder/Partner
Mathias Asberg
founder
Anastasiia Strakhova
Event Specialist at NewNative
Liza Marchuk
Dimitrije Pesic
Student
Omar Sesa
Assistant Lecturer in AI, ML and embedded systems field
This event has now ended, but you can still register for upcoming events on lablab.ai. We look forward to seeing you at the next one!
Checkout Upcoming Events →Submissions from the teams participating in the ChatGPT API & Whisper API Global Hackathon event and making it to the end 👊
In short, it’s a multilingual voice assistant, that can help Not only to reduce the language barrier in using cutting edge technologies but tries to make everyday life a bit easier. It also increases accessibility of the technology that can be helpful for people who have some kind of disability. With the use of ChatGPT & Whisper API from Open API and by using Google Translate library from Python and Text to Speech API from Google Cloud I have created a web app that can take voice input from user and perform multiple tasks based on User preference, such as language Translation and/or Communication with chatGPT to get information.
Zion
Our project aims to improve the academic performance of undergraduate students in their first cycles, helping them to adapt to the academic environment and develop effective study skills. To achieve this, we have created an innovative platform that integrates artificial intelligence technologies, such as GPT-4 and Whisper, to offer a series of tools that facilitate the learning process. The tools offered by our platform include automatic transcription of lectures, which allows students to access the information presented in a more accessible way; summary generation, which helps them understand and retain essential information from lectures; and personalized learning paths, which guide students to the most appropriate resources and activities for their individual needs. In addition, our platform provides research and writing assistance, which facilitates the completion of high-quality academic papers. All of this helps to reduce the stress and frustration associated with adjusting to the university environment and improve study skills, which in turn translates into better academic performance. In the $5.76 billion grade app market, our solution has great potential to revolutionize the way students deal with academic challenges. Moreover, being aligned with Sustainable Development Goal 4, our project also contributes to improving the quality of education and promoting academic success in a broader context. We are currently seeking support and investment to incorporate GPT-4 and Whisper into our platform and optimize it, thus ensuring that our solution is at the forefront of artificial intelligence technology and has an even greater impact on the lives of university students.
Smart Notes Learn Better Fast
The SaaS (Software as a Service) product is a powerful speech analysis tool that utilizes state-of-the-art speech-to-text technology to transcribe recorded audio and analyze it for various speech patterns. The tool identifies areas where the user may need to improve their speech, including reducing stuttering and improving clarity and coherence. Once the audio is transcribed, the tool provides users with in-depth statistics on their speech patterns, including word frequency, common mistakes, and areas that need improvement. These statistics are displayed in an easy-to-understand format that allows users to quickly identify their strengths and weaknesses. One of the most unique aspects of the tool is its ability to give real-time feedback on speech patterns. As users speak, the tool analyzes their speech in real-time and offers suggestions on how to improve their communication skills. This feedback is critical for users who want to improve their speech in real-life situations and gain confidence in their ability to communicate effectively. The tool also provides personalized suggestions for each user based on their speech patterns and areas for improvement. For example, if the tool identifies that a user tends to stutter on certain words or phrases, it will offer personalized exercises and techniques to help them reduce their stuttering and speak more fluently. Overall, the speech analysis SaaS product is an invaluable tool for anyone looking to improve their communication skills, reduce stuttering, and gain confidence in their ability to speak effectively. With its cutting-edge speech-to-text technology and real-time feedback, the tool offers a powerful solution for anyone looking to improve their speech and become a more effective communicator.
Coders Legion
Investing in the right talent is crucial to the success of any company, but finding candidates who are skilled and fit well with the organization's culture can be daunting. We have the solution. Our cultural fit bot, Cultivate, leverages the power of AI to help companies identify candidates who align with their culture, work style, and values. By automating the initial interview process, Cultivate enables recruiters to evaluate more candidates and make data-driven hiring decisions without compromising the quality of the hire. With Cultivate, recruiters and hiring managers can analyze candidates' responses to real-world scenarios and identify critical thinking, problem-solving, and communication skills. The bot also provides instant feedback on candidates' cultural fit, including their attitudes, beliefs, and behavioral patterns. It enables recruiters to identify top candidates that align with their company culture and values. Our platform continues beyond there. Cultivate also provides: An automatic interview summary for each candidate. Highlighting their strengths and weaknesses. A set of metrics that help recruiters make informed hiring decisions based on data, not opinions. These features significantly reduce the risk of bias in the hiring process, promoting fairness and inclusion. Investing in Cultivate means investing in the future of your company. By leveraging the power of AI to streamline the hiring process, we help you find top talent that not only meets the job requirements but also aligns with your company culture and values, ultimately improving productivity, employee engagement, and retention rates. Join us today and start growing your team with harmony and success.
Cultivate
We have created a no-code platfrom that caters to the 'customer care' department of the ecommerce websites, wherein the user can chat with our bot and the bot will figure out the problem/issue with the order and take action accordingly. Like for example, the user starts chatting with the bot and tells that the product does not fit his size, then the bot will traverse through the dataset and extract the order details and see if the product is elligible for return, if it is, then it will accordingly inform the user that the product can be returned and will place an order for the same. This no - code platform caters to the needs of ecommerce and similar online businesses that need 'customer care' service but cannot afford one. By means of this bot we are enabling that feature and making the platform no-code so that anyone can use it.
Clippy
SophistAI is a OpenAI, ChatGPT 3.5turbo based AI, that is aimed to make writing official documents easier. When you feel that pure use of thesaurus is just not enough… Create better impression of yourself by producing the most exquisite, most sophisticated sentences the English language has to offer! Just insert your humble text, to achieve the most astonishing result: "Keep your hair on – I only accidentally deleted your dissertation" And you`ll get a sophisticated sentence instead! “I apologize for any inconvenience caused, but please do remain composed as it was an inadvertent deletion of your dissertation." Below, you will find a version on google-collab and for discord.py. This project has a goal of helping journalists, reporters, detectives and businessmen or just anyone who would like to write with inspiringly beautiful language. Future plans: Switch to chat-GPT-4 Better offensive language handling Increase the maximum size for text inputs User miscellaneous controls Better sophistication of languages other than English
Sophistisolo
An AI-based audio to video converter is a tool that can convert an audio recording into a video file. This tool uses advanced machine learning algorithms to analyze the audio recording and create a video that matches the content of the recording. In addition, this tool can add text in terms of captions to the video based upon the speech in the audio recording. The AI system listens to the audio file and uses speech recognition technology to transcribe the spoken words into text. Then, it synchronizes the text with the video frames to create captions that appear on the screen at the appropriate times. The resulting video file can be used for a variety of purposes, such as creating video content for social media platforms, generating marketing videos, creating instructional or educational videos, and more. The AI-based audio-to-video converter makes it easy for users to quickly and efficiently create high-quality video content from their audio recordings,
Night Owls
ChattyRental is a revolutionary chatbot powered by AI, designed to simplify and enhance the room rental experience for both renters and rental companies. Our chatbot is equipped with advanced natural language processing capabilities, enabling users to book a room through a simple, conversational interface. Additionally, ChattyRental generates personalized recommendations for rental rooms based on a user's preferences and past behavior. With our intelligent search feature, users can find the right room that meets their needs quickly and easily. By leveraging the power of ChattyRental, room rental companies can cut their sales department costs by an average of 36%, while simultaneously improving customer satisfaction and user experience.
Room Booking AI Assistant
Have you ever wanted to test your knowledge on a specific topic, but found traditional methods of studying and taking quizzes to be tedious and unengaging? Look no further, because we have the solution! Introducing MyQuiz.AI, a trivia game that utilizes the power of AI to generate questions tailored to your interests and abilities. With just your voice, you can embark on a fun and challenging quiz journey that will leave you wanting more. So sit back, relax, and get ready to put your knowledge to the test with MyQuiz.AI! Our team has developed a cutting-edge speech-based game that incorporates advanced technologies to deliver a highly engaging and personalized user experience. The system utilizes the Whisper API for speech recognition, Redis for data storage, and ChatGPT to generate questions and validate user answers. The quiz asks unique questions every time, tailored to the user's level of knowledge and abilities. The system's machine learning capabilities ensure that the difficulty level of the questions is appropriate and challenging, and according to age. The Whisper API's advanced speech recognition capabilities provide an immersive and interactive experience, allowing users to use their voices to mention their age and category for quiz. This feature also makes the quiz accessible to users with disabilities or those who prefer voice-based interactions. The Redis database stores questions, answers, and user responses. Overall, our speech-based quiz or game represents a significant step forward in the field of educational technology. With its advanced algorithms and machine learning capabilities, the system offers a new and innovative way for users to learn and engage with the material. The quiz's personalized approach, speech-based interface, and advanced features make it a powerful educational tool that has the potential to revolutionize the way people learn and retain knowledge.
Space Cats
Our product takes in a job description and helps prepare a user for a job interview. The model uses the job description searches through a database to find the most relevant documents to give chatGPT context. Then it provides them interview questions. After the interview, which uses the Whisper API to transcribe the user's voice response to the interview question, the model provides feedback by looking using the question and the user's response to find relevant documents to give ChatGPT context on what makes a good answer. After the user has completed all interview questions. They are given overall feedback, and a final score on their performance.
Worktern
Our app is designed to address some common problems that students and learners face when trying to engage with lectures Difficulty taking comprehensive notes: Many students struggle to capture all of the key points and details of a lecture while also actively listening and processing the information being presented. This can result in incomplete or inaccurate notes that make it harder to study and review later. Time-consuming manual transcription: In order to review lectures later, students may need to manually transcribe the audio recordings, which can be time-consuming and tedious. Limited ability to identify important information: Even with comprehensive notes or transcripts, it can be challenging to distill the most important information from a lecture, especially if there is a lot of extraneous detail or repetition. Our app aims to address these problems by automating the process of creating summaries, notes, and questions from lecture audio. By using WhisperAI to transcribe the audio to text and ChatGPT to generate a summary, notes, and questions, the app streamlines the process of reviewing lectures and helps learners more easily identify and retain key information. Here is a possible flow for the app: The user opens the app and selects the lecture they want to review. The app uses WhisperAI to transcribe the lecture audio to text. The text is passed to ChatGPT, which generates a summary, notes, and questions based on the content of the lecture. The user can review the summary, notes, and questions generated by the app, edit them as needed, and save them for future reference. Overall, this app has the potential to be a valuable tool for learners who want to optimize their engagement with lectures and maximize their retention of important information.
The bad batch
MyCademy is a revolutionary AI model that redefines the online learning experience. By leveraging the power of cutting-edge technologies like GPT and DALL-E-2, MyCademy creates a fully personalized course for each user based on their unique needs and interests. Not only does MyCademy generate high-quality text-based content, it also uses DALL-E-2 to create stunning and immersive images that enrich the learning experience. But that's not all! MyCademy also includes a quiz feature that enables users to test their knowledge and track their progress. And for those who want a quick summary of the course, MyCademy provides an easily digestible synopsis that ensures maximum comprehension. The interactive chatbox feature is where MyCademy really shines. Users can ask questions and get immediate answers, allowing for a truly interactive learning experience. Whether you're struggling with a concept or looking for additional information, MyCademy's chatbox is always available to help. In short, MyCademy is the ultimate tool for those who are serious about online learning. With its personalized course creation, immersive images, quiz feature, and chatbox, MyCademy makes learning more engaging, interactive, and effective than ever before.
The Cyber Savvy Ninjas
an innovative and user-friendly health application that uses artificial intelligence to provide Ugandan citizens with access to vital health information. The app is designed to address the challenges that many Ugandans face in accessing quality healthcare services, particularly in rural areas where health facilities are scarce. The application is built using state-of-the-art technology, including ChatGPT API for natural language processing and speech-to-text capabilities, Streamlit for the user interface, and the Reddit API to access relevant health information. These tools work together seamlessly to provide a comprehensive and user-friendly health platform that meets the unique needs of Ugandan citizens. Through the app, users can access reliable and up-to-date information on common illnesses, including symptoms, causes, and treatments. They can also receive personalised recommendations based on their symptoms and medical history, as well as find nearby health facilities and book appointments. The app can also provide educational resources on topics such as sexual health, maternal and child health, and HIV/AIDS. The app's user-friendly interface and speech-to-text capabilities make it accessible to all Ugandan citizens, regardless of their level of education or literacy. This is particularly important in rural areas where illiteracy rates are high. Additionally, the app's use of local languages such as Luganda and Runyakitara ensures that it is inclusive and accessible to all Ugandans. Overall, "Health Solutions Uganda" is a powerful tool that has the potential to revolutionise healthcare in Uganda by providing access to vital health information and services to all citizens, regardless of their location or socioeconomic status.
BadaRama
QuizTube is a web application that generates multiple-choice questions based on the audio content of a YouTube video. Users can enter the link to a YouTube video, and QuizTube will download the audio from that video, use the Whisper API to transcribe the video then submit a request to chatGPT to generate multiple choice quiz based on the content in the transcription. The questions are designed to test the user's understanding of the content, and the app can be used for educational purposes, language learning, or just for fun. With QuizTube, users can turn any YouTube video into an interactive quiz!
chatgpt5
We have built a solution for agencies which provide the caretaker services for parents who are in search of babysitters for their child. When users call the agency after business hours or when agents are not available for assistance, we are routing them to leave a voicemail with their babysitter requirement and contact number. With this solution, agents can focus on more complex tasks rather than manually retrieving voicemails, analysing them and coming up with a resolution. When the caller dials the agency phone number during office closed hours or peak hours when agents are not available to serve them, we route the caller to the voicemail menu where we ask them to leave a voicemail with babysitting requirements and their contact details, etc. Once the voicemail is available, we extract it and convert this speech to text using OpenAI’s whisper API which gives us the voicemail transcription. After that, we meticulously perform the prompt engineering for ChatGPT API to provide us all the required information from voicemail like intent, sentiment, babysitting date and time, etc in JSON format. Using this information, we query the EmployeeSchedule table which is in the H2 database. Once we have the information about availability of babysitters, we query RedisJSON to get the employee profile information like employee name, contact details, date of birth, languages spoken, image, etc. We then build a PDF document using itext library. This PDF containing available babysitter information will be sent on the caller’s WhatsApp. After this, we send an SMS to the agency as an alert notification about the customer enquiry and ask them to get in touch with the customer. Github link - https://github.com/technocouple/technocouple-caretaker-assistant Video link - https://drive.google.com/drive/folders/1NBew2U0Xgtm04ubQszjLvZV92fowR6-D?usp=sharing Presentation - https://drive.google.com/file/d/1TBMSU5Ohyn1v2P2u_RqbZOpuCvWv1Crq/view?usp=share_link DEMO is at the end of the video.
TechnoCouple
In a world with so much information and with so much things to do in a single day, we need the best information fast at anytime we want to. What if there was an AI Chat bot to talk about sexual health, pregnancy and parenthood that knows everithing you need and can recomend content and experts to access everytime you need. Kuna life will be a platform to take care life and humans with trusted knowledge, we want that you can fill calm and don´t waste your time trying to find the right answer with the help of AI we could bring the solution in an easy way, we want to take care of you and your family.
KUNA
When you're interviewing, it's important to focus on the process and listen carefully to the interviewee and get into the process. But when you're constantly distracted by taking notes and looking at a list of questions, you lose your effectiveness and maybe forget to ask something. Our app is designed to save the interviewer from unnecessary activities and help him or her focus on what's important. Now the interviewer can take notes and try to write down the interview, because our app will do it for him. It will also review the entire resume and answer specific queries. This is just the basic functionality we managed to implement in 48 hours. In the future, the app can work in real time, toss questions and give hints, also will save the processed interviews
Nathnenne
Curated Club is a subscription-based service that offers monthly deliveries of curated products related to the customer's individual interests and preferences. The service uses a personalized algorithm to analyze customer data, combined with ChatGPT API to understand natural language, and customer feedback to curate a selection of high-quality products that are tailored to each customer's specific needs and preferences. The service offers a wide range of themes to choose from, such as food and snacks, books, pet care, fitness, and more. It is designed to offer a fun and convenient way for customers to discover new products and hobbies, while also providing a personalized and seamless experience that keeps them coming back for more.
TalkyAI
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.
UncomplicateIT
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.
Dripper News
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.
MASTERS
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.
Pac-Man
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.
New Era
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.
Solorider881909
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!
ProjectK AI
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
Tengri AI
Call centers handle an immense volume of customer interactions every day, 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. This approach is prone to human error, inconsistency, and scalability challenges. The Voice Analytics with AI aims to revolutionize the quality & assurance auditing process of call centers by transcribing and analyzing audio recordings using GPT-3 and Whisper models. The proposed solution leverages Automated Speech Recognition (ASR) and Large Language Models (LLM) to automate and streamline the quality and assurance auditing process. First, the system summarizes key information in call recordings, such as operator's name, issues, and solutions, and other relevant data points. After that, the solution conducts sentiment analysis to evaluate the tone and mood of the conversation using NLP and LLM. In addition, LLM evaluates customer experience and satisfaction levels and provides scores for each. Last but not least, the model ends the report of each call with feedback and insights about the performance of operator and suggests areas for improvement. Overall, the proposed solution has the potential to transform the call center industry, providing businesses with valuable accurate insights into their customer interactions and enabling them to take proactive steps to train their operators and improve their overall customer experience.
AI Alliance
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.
MagnaLingua
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.
GptHelpLine
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.
team phoeniks
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.
Mentorsko