Speech to text components

Speech to text components enable a voice assistant to recognise user's speech and format it to the plain text. Then this text should be passed to another component named Dialog API to recognise a user's intention.

Built-in speechkit libraries

Aimybox provides built-in speech to text components available via library dependencies:

• Android speechkit (supports offline mode)
• Kaldi speechkit (supports offline mode)
• Pocketsphinx speechkit (supports offline mode)
• Google Cloud speechkit
• Houndify speechkit
• Yandex Cloud speechkit

How to add another speech to text engine

You may wish to add another speech to text engine library for your assistant. To do this please implement SpeechToText. You can see how it's done in any of built-in engines listed above.