The new free service Otter will surely interest those who, by the nature of their work, have to record spoken language and then convert it into printed text. It was presented at the Mobile World Congress 2018.
Such free applications are no longer a novelty. Their common lack - a lot of inaccuracies in the compilation of the text. In turn, good, quality applications are very expensive. From them Otter is distinguished by the presence of artificial intelligence, which converts speech in real time. For the synthesis of artificial speech, perspective algorithms have already been created, for which Otter is oriented. One of them is Google DeepMind, which is remarkable for its accuracy and clarity.
Otter is easy to use. The application starts working with the "Record" button. As soon as the user starts talking, a draft of the text version appears on the smartphone's display. After the completion of the statement, the artificial intellect enters into the matter, which edits the text.
The application has a useful feature. During the interview, Otter identifies both of its participants, separating the text into understandable replicas. To do this, the interviewer should leave a "voice print" before starting the dialogue, having read several suggestions on the microphone.
The drawbacks of the application include difficulties with the arrangement of punctuation marks, the problem of working in crowded places or in rooms where there is much noise, but its developers promise to eventually bring it to "mind."