geigster

We're in beta so it's free. But we'll monetize by keeping your data safe, and offering premium services. Not selling it to ad networks :)

Nice!

Export feature this week, Zapier really soon, API later this year.

We want to be your interface to create and validate, so we're going to help get your data out and where ever it needs to go.

One day soon we'd like for you to be able to build your own workflows, validation and enrichment rules. That's a little while off, but that's where we're heading.

They sure do! You can tag anyone, not just users!

We use a major platform, and our customer data is not shared to train their system.

It limits our flexibility a bit, but it means our customer's data isn't being farmed off to a tagging factory somewhere.

No small startup can afford to do more than 1 or 2 languages, infrastructure cost overhead is incredible and the platforms keep moving faster and faster.

We think of these engines as cloud services, we're productizing infrastructure.

Ouch.. you're right. Thanks for flagging it. The new page went up over night, and we missed this.

The team is gonna get on that. No good excuse for heavy graphics, that's not how we roll.

You nailed it! Internally we don't really assign dates either, but certain user profiles like the idea.. if not the practice :)

One of those marketing features- like buying a plane ticket with Siri.. Cool idea, never happened :)

We plan to get a lot smarter with enrichment in future - like referring to docs or projects to bring them into the conversation.

Thanks for the feedback! It's a new concept and we have a ways to go to get better at explaining it. Easiest to just try :)

We're experimenting with dictation that keeps the audio, allows you to format it in a way that can be read (or used by other systems... like project management tools or CRM). You built up notes ad tasks with lots of little grabs. Like writing, you talk in bursts.

Keeping the audio means you don't have to worry as much about transcription errors or misunderstood tone. You can be very fast and casual - unlike written - and the audio is a safety net.

Voice recording takes a little getting used to, a lot of execs currently record voice memos for their EA... who then have to type it into something useful.

We hope to make this behaviour accessible to everyone, because it's fast and kind of amazing when it works.

This is a great idea, we hadn't thought of that. We speak 6 languages in our team, so its come up a lot.

We played around with passive detection, but it's still too flakey and ASR breaks down entirely if you start pronouncing names with a different languages pronunciation.

English - Hugo - fine! English - Portuguese pronunciation - Ugo - 100% fail. Portuguese - Ugo - fine!

Awesome feedback!

Thanks @rekoros!

I'm one of the makers, happy to answer questions!

Haha :) But then you have to stop and listen... and write it down somewhere :) ASYNC FTW.