Here is an artificial intelligence tool which can segment speech into words or morphemes to pre-process the audio materials and help the transcription work. Right now this tool is at the stage of being tested and improved, because currently the most urgent task to enhance its accuracy is to collect enough suitable samples for its training process. So feel free to have a try! We need your valuable sample to improve our accuracy. This machine has the advantage of its relatively good versatility: as long as the language belongs to the Sino-Tibetan language family, especially a Tai-Kadai, Sinitic, or Tibeto-Burman language full of lexical tone changes, basically it can be segmented well in this machine. (Note that the tool only reads audio data in wav format.)

No comments have been posted yet.

Sign in to leave your comment