The new free service Otter will surely interest those who, by the nature of their work, have to record spoken speech and then convert it into printed text. It was presented at Mobile World Congress 2018.
Such free applications are not new for a long time. Their common drawback is a lot of inaccuracies in the composition of the text. On the other hand, good, quality apps are expensive. Otter differs from them in the presence of artificial intelligence, which converts speech in real time. For the synthesis of artificial speech, promising algorithms have already been created, which Otter is focused on. One of them is Google DeepMind, which is amazingly accurate and clear.
Otter is easy to use. The application begins to operate with the press of the "Record" button. As soon as the user starts to speak, a draft of the text option appears on the smartphone display. After the end of the statement, artificial intelligence comes into play, which edits the text.
The application has a useful feature. When conducting an interview, Otter identifies both participants by dividing the text into understandable cues. For this, the interviewer must leave a “voice imprint” before starting the dialogue by reading several sentences into the microphone.
The disadvantages of the application include difficulties with the placement of punctuation marks, problems of working in crowded places or in rooms where there is a lot of noise, but its developers promise to bring it to the "mind" over time.