Yandex SpeechKit

A service that recognizes and synthesizes speech in multiple languages.

Used by Yandex’s Alice voice assistant, adaptable for use in your business solutions.

Get started Documentation

Context-based recognition

SpeechKit takes into account the probability of word combinations and lexical and stylistic features of oral speech.

Real-time synthesis

Text synthesis is done with minimum delay and works perfectly in streaming services.

Support for three languages

The service handles audio and text in three languages: Russian, English, and Turkish.

Premium voices

Premium voice speech consists of a million individual phonemes and sounds natural. Before starting speech synthesis, the service evaluates the entire text and selects the intonation characteristic of human speech.

Transparent pricing

The cost of audio recognition is automatically calculated based on the length of the track. The cost of text synthesis is based on the character count.

Implement your projects using SpeechKit

Automate info recognition and handling for recording customer calls. SpeechKit recognizes speech, including the last name of the caller, preferred date and time of the appointment, and other details. Let your call center staff focus on more complex issues.

Give users the same information, referring to each person by name or a different ID during the call. Speech synthesis technologies help you personalize your message without involving a call center staff.

Add voice control to your app: it’s fast and convenient. Yandex SpeechKit can decode voice commands so that the app can respond to them.

Customer stories

The main reasons for migrating to Yandex Speechkit for speech recognition were its ease-of-use and convenience. These let Igor Druzhkov, our lead software engineer, quickly integrate our accounting system with Yandex Cloud.

Alexander Popov,

Head of IT department at Gazprom Mezhregiongaz Ukhta

Read the story

Questions and answers

How do I use SpeechKit?

The service runs via the HTTP API. You can find all instructions for using the service in the documentation. Get started with the service yourself or contact us. We’ll select a partner that will develop a solution specifically for your project.

Documentation

Why register in the Yandex Cloud console?

To use the API, you need to get an ID (an IAM token or API key). This ID is linked to your account in the cloud.

Documentation

To use the API, you need to get an ID (an IAM token or API key). This ID is linked to your account in the cloud.

Documentation

What is a recognition model?

A recognition model is a neural network that is trained to recognize speech in a specific language. The models are trained on datasets generated by Yandex services and applications. This allows us to continually improve the quality of speech recognition.

Documentation