Yandex Cloud has updated its service for speech synthesis and recognition based on machine learning Yandex SpeechKit. Now he can place punctuation marks when translating voice into text. This will help in areas where a person must interact directly with speech technologies. For example, in communication with a voice assistant, automatic transcription or subtitle formation.
The new feature is called Punctuator. It can work in real time for scenarios with voice assistants and when recognizing pre-recorded audio files. The punctuator was developed using two consecutive machine learning models. One translates the voice into text, the other places punctuation marks in accordance with the norms of the Russian language.
The function is available to all Yandex SpeechKit users and is configured using standard tools in the API.
Vasily ErshovHead of Yandex SpeechKit Development
“We develop Yandex Cloud services by responding to the requests of our customers, and sometimes even anticipating them. Many of our customers today use Yandex SpeechKit to create voice assistants with a conversational interface. In such projects, it is important to make the text recognized by the machine as close as possible to how a real interlocutor would write it. The punctuator solves this problem simply.
bbabo.Net