EXMARaLDA and Automatic Speech Recognition (ASR)

Automatic speech recognition (“Speech-To-Text”, ASR) has made significant progress in recent years. In some scenarios, it can now replace manual transcription or at least complement it to increase efficiency. The latest EXMARaLDA preview takes this into account by providing new and revised import functions for formats commonly output by ASR systems. On the one hand, these are the SRT and VTT formats, which have their origin in the subtitling of videos. On the other hand, the EXMARaLDA Partitur-Editor and FOLKER can now also directly import the JSON formats written by Whisper and by Amberscript.