Wearable Signal Tools · LLM agent tooling
TimeCap
deterministic captioner for wearable health time series
What is TimeCap?
TimeCap is an open-source, deterministic pipeline that turns multi-channel wearable health time series into text captions, with no language model in the loop. Statistical, structural and semantic extractors plus a cross-channel synthesis layer write the captions, and a new data schema needs only channel definitions. It is used to build text and time-series training pairs for sensor-language models, with adapters for MyHeart Counts weekly data and a WESAD demo.
TimeCap at a glance
- Signals
- Multi-signal
- Category
- LLM agent tooling · also preprocessing
- Language
- Python
- Licence
- MIT
- Commercial use
- yes
- Maintenance
- Commit in the last 12 months (as of Oct 21, 2026)
- Last commit
- Oct 14, 2026
Related open-source tools
Other Wearable Signal Tools that share a signal with TimeCap.
Summary, licence and commercial-use fields are shown as recorded for this project. Maintenance is computed from the last commit date. Check the project’s own repository for its current licence and status.