Large number of creators, experts, professionals are publishing their words in form of recordings. They provide valuable content in the form of unstructured audio signal that is difficult to approach, read, search and obtain valuable insights.
We create a tool for automatic speech-to-text conversions optimised for closed captions and written transcription.
We train our machine learning models on large portion of data that we've gather and enlarge constantly. With such approach we target slavic languages and creators within Central-Eastern Europe.