We do have an own speech to text solution whcih we do want to integrate. Is there any documentation on how to integrate such a custom module into mycroft?
Thanks for your reply. I do know this page in the documentation. There is all the information available if i want to use an available STT engine like google, kodi, amazon, …
But I do want to use my custom/self-made STT engine. Is there any documentation on thow to build such a custom integration? Or any advice where to look at?
you can make a plugin
you can check out my plugins for reference
Look into https://github.com/MycroftAI/mycroft-core/blob/dev/mycroft/stt/init.py to get an idea how the existing STT modules were implemented. The basic principle is that the recorded audio is passed as file/stream to the STT engine and a text-string is returned. Everything else is engine-specific checks for supported languages, credentials, audio format etc.