Screenshot

Inspiration

In a world that thrives on communication, language barriers persistently stand in the way of personal and professional growth, cultural exchange, and global understanding. With over 7,000 languages spoken worldwide, the ability to learn and communicate in a new language has never been more critical. Yet, traditional language learning tools often focus on reading and writing, neglecting the crucial aspect of spoken conversation. This gap inspired me to develop "Speak" during the GHW - an AI-powered app aimed at bridging this divide by facilitating spoken language practice in a natural, interactive manner.

What it does

"Speak" enables users to engage in simulated conversations with AI, practicing spoken language in a dynamic and responsive environment. By leveraging state-of-the-art AI technologies, "Speak" offers real-time voice transcription, conversation simulation, and personalized feedback, including corrections and suggestions. This immersive approach helps learners improve their pronunciation, fluency, and confidence in speaking a new language.

How we built it

Built as a Python-based application, "Speak" integrates various AI and machine learning technologies:

Frontend: Developed with Streamlit, providing a user-friendly interface for real-time interaction.
Voice Transcription: Utilizes Whisper for accurate, real-time voice to text conversion, ensuring learners' inputs are precisely captured.
Conversation Simulation and Corrections: Employs Hugging Face and Ollama for generating realistic and culturally relevant dialogues, along with language correction capabilities.
Voice Cloning: Incorporates the Bark model for voice cloning, offering personalized feedback to enhance the learning experience. This multi-faceted approach allows "Speak" to offer a comprehensive and effective language learning tool. ## Challenges we ran into One of the main challenges was to create an intuitive user experience that would appeal to both those in the know and those less familiar with AI technologies. Meeting these challenges required innovative problem solving and extensive testing to refine the application's functionality and usability.

Accomplishments that we're proud of

Successfully integrating complex AI technologies to simulate realistic conversations.
Developing the application as a solo project, which involved managing every aspect from conception to implementation.
Creating a scalable, open-source platform that can continuously evolve with community input.
Receiving positive feedback from early users on the significant impact "Speak" has had on their language learning journey.

What's next for Speak

The future of "Speak" is bright, with several enhancements on the horizon:

Cloud Deployment: Planning to containerize "Speak" using Docker, facilitating easier deployment and scalability.
Language Expansion: Aiming to support more languages and dialects, broadening the app's accessibility.
Community Building: Focusing on fostering a community of learners and contributors who can share resources, experiences, and improvements. "Speak" is more than just a language learning app; it's a step towards a more connected and understanding world.