found in translation

Inspiration The inspiration for "Found in Translation" came from the desire to create a solution that seamlessly bridges language barriers in an increasingly interconnected world. By leveraging the portability of the OMI AI wearable device and the robustness of blockchain technology, we sought to empower users to communicate effortlessly in multicultural settings while maintaining security and privacy.

What it does "Found in Translation" integrates real-time speech and text translation into the OMI AI wearable device. The application leverages Polygon’s on-chain infrastructure to manage data securely and Phala Network’s tools to ensure privacy-preserving computation. This allows users to engage in fluid, multilingual conversations without worrying about data misuse or accessibility, making the device practical for global travelers, professionals, and anyone navigating diverse linguistic environments.

How we built it The project combines advanced AI models for natural language processing with decentralized technologies. Using Polygon’s blockchain for secure and scalable data handling, we ensured that translation activities could be stored and processed transparently. Phala Network’s dstack was employed for edge computing to protect user privacy while enabling fast, real-time translations. The wearable device was designed to be intuitive and lightweight, ensuring that users could engage with it effortlessly. ETHGlobal’s collaborative environment provided valuable feedback, which helped refine the project further.

Challenges we ran into We faced significant challenges in maintaining translation accuracy and responsiveness while working within the hardware limitations of a wearable device. Ensuring seamless on-chain integration without introducing latency was particularly difficult, as was achieving a smooth user experience during extended conversations. Balancing privacy and usability with Phala Network’s agent contracts required innovative solutions to ensure robust performance on a compact platform.

Accomplishments that we're proud of Winning 2nd place for Best AI + Onchain App on Polygon at ETHGlobal validated our ability to combine cutting-edge AI with decentralized technologies. Additionally, securing 1st place for Best Use of dstack or Agent Contract with Phala Network showcased our innovative approach to privacy-preserving, edge-based solutions. These recognitions highlighted the practical and technical impact of the project and its potential to redefine wearable technology.

What we learned We gained valuable insights into the flow of language translation during dynamic conversations, particularly when extended interactions occur. In scenarios where the conversation is longer or involves multiple speakers, we discovered that having more than one OMI wearable device can significantly enhance the experience. This ensures that each participant can engage fully, without relying on a single device to manage complex multi-party dialogues. Additionally, we learned the importance of maintaining contextual awareness during translations to ensure that the flow and meaning of conversations remain intact.

What's next for Found in Translation Our next steps involve enhancing the contextual understanding of the translations to support nuanced and culturally sensitive interactions. We aim to expand the device's capabilities by adding more languages and refining its on-chain features for identity and data management. Future iterations will also explore the potential for multiple OMIs to synchronize during conversations, creating a networked translation system that further enriches the user experience. Collaboration with partners like Polygon and Phala Network will remain a cornerstone as we scale the project and introduce it to broader markets.

Built With

• amazon-web-services
• dstack
• edge-computing
• flask
• ipfs
• javascript
• mongodb
• node.js
• openai-gpt-api
• phala-network
• polygon
• python
• react
• webrtc-api