Gradio
Build interactive machine learning interfaces with Python
FUTO Swipe
Build on-device swipe typing with open models and an inference library
Meet FUTO Swipe
FUTO Swipe is a collection of models and algorithms for recognizing words from swipe gestures on a keyboard. It powers swipe typing in FUTO Keyboard for Android and is also available to developers building other input interfaces. The system combines gesture recognition, language context, and dictionary-based decoding to turn a path across keys into candidate words.
Developers can obtain the models, a C++ inference library, and a published dataset for experimentation or integration. The production system runs on the device, while the public webpage provides a server-backed demonstration. Its language-specific decoder currently focuses on English QWERTY input, with separate licensing terms for the models, library, and dataset.
Discussion
Tried it? Share a useful thought or ask a question.
Good conversations start with a hello.
Sign in to commentProduct details
- Website
- swipe.futo.tech
- Status
- Launched
- Category
- Developer tools
- Listed
Explore related topics
Similar products
Lloyal
Build applications with local models and agents inside the runtime.
Kotlin
A statically typed language for Android, backend, and multiplatform applications
slash-editor
Add an editable block interface to a React application
Java
A programming language and platform for portable applications
Files SDK
Use one file-storage API across cloud providers and local backends