2 ms·
Anyone have any experience with it? How accurate is the video? For example, second use case - in the car, she queries the software the amount of some_specific_
by komali2 9y ago
Anyone have any experience with it? How accurate is the video?
For example, second use case - in the car, she queries the software the amount of some_specific_product in some_specific_warehouse, and the tool answers near instantly. Is that a currently functional scenario? Is this because the client has an API set up with `counts` already calculated for these queries, or is a part of the tool intelligently crafting these queries to answer these kinds of voice questions?
- mpeg 9y agoFrom the article, the launch product is simply a customer service chatbot (currently text, with voice coming soon) The omega product video is supposed to be the future vision and not a real product.
- Eridrus 9y agoNo insight into Omega, but natural language interfaces to spreadsheets/databases exist. Google Sheets has an implementation if you click the explore button in the bottom right hand corner. I made a dummy spreadsheet with columns name & warehouse, then typed in "how many units of rose gold are there in la" and it got the query right. It even got "how many rose gold units are in los angeles" correct despite the spreadsheet never saying los angeles. But it's pretty easy to trip it up; if you make an item called "rose gold" and "rose gold ring" and ask it for rose gold, it assumes you want the exact match. If you ask it for items in la or sf, it will only give you a count for the first condition. If you ask it about California, it will get confused. Still pretty impressive for a two column spreadsheet though IMO.
- dagw 9y agoAssuming the system is designed to intelligently integrate with SAP and other ERP systems then it shouldn't be too hard. The trick (and what people might actually pay for) is making this integration work.