Question about MLC output latency on the LSM6DSV16X

[Annika Jiang]

Hi everyone,

We’re currently evaluating the LSM6DSV16X for a multimodal data acquisition application and are using its Machine Learning Core (MLC) to detect specific events and generate corresponding labels in a time-series data stream.

To ensure accurate temporal alignment between different data sources, we’d like to understand the latency between the actual occurrence of a physical event and the availability of the corresponding MLC output.

Has ST characterized this latency or published any relevant specifications, test results, or documentation? If not, is there a recommended test procedure or validation methodology for measuring it reliably?

Any insights or suggestions from the ST team or community members with experience using the MLC would be greatly appreciated.

Thanks in advance for your help!