Let's let's talk a little bit about search invoice. So that's a feature that l. Kolia added a couple of years ago and in your perspective on this having been in this search field in the past when you think about any type of natural language input that's different than structured input. So that's the first base and then when we talk about natural language input typing versus speaking. We also have some differences there so be interesting you talking about sort of that spectrum. And maybe there's more elements to. I may be oversimplifying it. A little bit. But i'd be interested in few elaborating that and search because it's so important to the future efficacy of all these voice applications that people are making yes absolutely and so the interesting thing about Voice search if you can trust what you're i think. User expectations are one of the biggest. Starting points of difference. So when you're when you're searching for something via voice it's not a question of returning the top. You know for results and letting somebody choose. That's not an option. It's you know you have to get to the right answer. The first time so contents context is king. So you know again. It's probably the personalization journey. The more you can know about somebody the more you can get it right for them and if you can do that in the moment then then even better. But i think there's an the element of having to get it right the first time a lot of pressure on the system But that's exactly what you need to be able to do with voice. I think the other aspect to this is the combination of voice. Plus you know the chat bought in the bottom right hand corner if you like the screen. I think there's a lot of opportunity for I think there's a lot of opportunity for natural language search voice such when it comes to customer success or customer service In a retail type of a scenario or any type of scenario you know. I had a problem last week with a a delivery. I choose a meal service and i have a lot of. I'd have facetime ever. I've been using them but they might delivery got lost And you know say that was. Ups problem was you know passing the buck. I needed to solve enga delivery so i was doing it via the chat ball but it was really hit and miss it was. You could tell it wasn't it was just not a good experience whereas had that been far richer at natural language type of such that they were using and made conversational. Then that would have been such a better customer experience for me. So there is so many I think different use cases for this that. It's you know the possibilities are endless..