Hard work pays of. As mentioned in our last newsletter, we have been working on updating our API and now we are proud to say we’re live! We’ve made many changes ‘under the hood’, but updating the interface was long overdue as well. Let me walk you through some of the key improvements that enable you to use our API in a more easy and overall better way than before.
Interface; from rags to riches
Initially for our first version API we purely focussed on having a functional system running, allowing our users to transcribe content in real-time and batch mode. And to be honest, developers – who mainly use our API – aren’t that interested in a flashy interface anyway. However, a lot of people who want to try out our speech recognition models need to be able to work with an easy to use interface.
Register and start testing
Now everyone can simply register and login. No need for any technical knowhow, everybody who registers receives one hour of quota per language for free! Simply upload a file in any of the languages we support and check the transcript results. We’ve visualized the confidence score our engines add to all words in a transcript, allowing you to run some tests and compare to other vendors before choosing to actually onboard.
Improved usage insights
Additionally we provide customers with a better usage overview which they can cross reference with invoices they receive from us. You can filter for a specific period in time and download the results in csv for your own administration.
Create tokens and retain
If you need multiple tokens, for your own application(s) or for your customers when reselling our speech recognition models, simply create a new token key for the service you need “batch” or “real-time” mode, add a name to your token and validate. Now go ahead and use your newly created token to transcribe your audio content for yourself or your customers.
Any questions about our new API?
Get in touch and we’re glad to help out!