Speechly (https://www.speechly.com/) enables you to create new kind of voice user interfaces with real time visual feedback easily for any use case.
Easy to use libraries available for iOS, web, React and Android. Great use cases:
- any app that needs to input (modifiable) items, such as food/health tracking, fast food ordering, etc
- any app that has search filtering (see fashion.speechly.com for a demo)
- or of course anything that benefits from voice, such as innovative games, apps used on the go etc
Check it out!
Here's one case where I clocked an over 3x improvement when measuring time only. Probably would go as a 10x improvement in some cases (accessibility, intuitivity etc)
https://www.youtube.com/watch?v=Q2gwzTWADns
Hi there! Voice search can of course be done by using something like Amazon Transcribe or Google Speech to Text. The problem however is that they just turn speech into text and that's not the best way to leverage voice in search.
With these tools there's a bit jarring experience with waiting for the results and then either you have to submit the search query automatically and trust that there was no mistakes with the speech recognition or you have to click enter after making sure it was understood correctly (editing wrongly recognized utterances especially on mobile is very frustrating)
While these are small details, they make the experience a lot worse.
Our tool (yes, we have a technology that can be used here) combines speech recognition and natural language understanding in a novel way and in fully streaming fashion. It enables search experiences like this https://www.youtube.com/watch?v=xI68NT8D1m8 where the UI is updated as the user speaks. This makes it easy for the user to correct themselves by using voice and it encourages them to go on with the voice experience if there's no errors.
Our tech is applicable to web very well: works on all platforms and we have client libraries for React and JavaScript for easy integration.