Android system settings for speech and voice recognition

  • You may need to tweak your system settings for an optimal experience. Here are some tips for setting up Android for speech and voice recognition.

    Speech Recognition (Speech to Text):

      Speaking Email uses "Google Voice Typing" for dictation and commands. It works offline so you need to download the offline voice.

      When you have done this, you need to make sure the same language (eg en-US, en-GB, fr-FR, pt-BR) is used in various settings so that everything works.

      Check the following in your system settings under Language & Input:

      1. Look under 'Language & Input'. 
      2. Find "Google Voice Typing", make sure it's enabled

      3. Tap 'Offline Speech Recognition', and install / download all languages that you would like to use.
      4. Note: if your preferred language does not have an offline voice available under Google Voice Typing, you must choose a language that does and make it your default language. Go back to 'Google Voice Typing', and select the languages you want to use under 'Languages'.
      5. Go back to 'Language & Input', and select the same languages again. Then select your primary language. Note that this must be one of the 'Offline Languages' you downloaded and installed first. A common problem is that the default language is not available as an offline voice.  For example en-NZ is not available offline so you must use en-US or en-GB as your default language (but you can keep en-NZ as a secondary language)

      Still not working?

      You may have some system setting inadvertently disabled. The best way to check is to open Google Voice Assistant. If something is disabled it will detect and prompt you guide you how to turn it on. 

      Speech Output (Text-to-Speech aka TTS)

      • Look under Accessibility > Text-to-speech output. Ensure you have "Google Text to Speech" selected and the correct language. 
      • Note that Speaking Email won't use Samsung or other vendor voices - so you need to enable the Google voices as your default TTS engine.