Do you want to save time, improve your workflow and be able to search within audio visual content? All this is possible with our Speech to Text service. We transcribe all your audio and video’s in real-time and synchronize the audio to the transcript. But there’s more…
Select a quote or word within your transcript and you’ll see the exact time it was mentioned in the recording as well as the duration. This can be a huge time-saver for anyone who needs to edit audio or video.
We’ve added voice recognition so you can search for specific people in your files. This makes it easy to search through great amounts of data, such as video archives or a podcast database.
We guarantee a minimal accuracy rate of 90 percent for all the language models we provide. However this is just the beginning, because we keep improving our language models on a daily basis. We integrated a human ‘check’ in our workflow, which we call the Human Feedback Loop. It’s pretty easy. We check the transcripts and if we notice any mistakes we’ll manually correct them and retrain our system with the corrected version. We do the same with a dataset for new clients. We collect all kinds of words that are important to them, such as product names or industry specific language. We add these words to our database and train our system with real (audio) content. We don’t believe in plug and play. We’ll keep improving and adjusting our software so it matches your needs.
It doesn’t matter if you’re a podcaster making one hour of content a week or a broadcaster with hundreds of videos. We work together with Microsoft Azure to ensure you a safe and scalable service. We don’t store any of your content and use secure streams so you don’t need to worry about any privacy issues.
Would you like to test our Speech to Text service? Register and try our API for free.