Device voice, recorded drafts and Quick Voice

For: Student, Teacher, Centre Manager, Super Admin

Look for: Speak to my AI tutor · Record a spoken draft

1. Listen with device voice is always available and reads displayed text using an installed voice when the student presses Listen. It does not start automatically or use the AI budget. A matching local voice must be installed; the app does not silently send text to a cloud voice when one is missing.
2. AI voice is an optional, separately labelled choice enabled by an authorised teacher or admin in Students & plans → the learner → Language, voice and reading support. An admin override takes priority. The centre external-voice switch also applies. Saved AI recordings use these permissions but replay without a new AI call; new speech generation and transcription also use the existing eligibility and budgets.
3. Record a spoken draft to practise your answer. Start, speak clearly, stop and replay it. When cloud transcription is allowed, transcribe and check the words before using a formal answer.
4. Quick voice: speak and send is for informal tutor chat. Clear speech is transcribed, sent to the text tutor and can receive spoken output. Unclear words or website-like text require confirmation.
5. A microphone icon and sound meter help identify recording controls. Allow microphone permission, speak close enough and stop when finished; recordings stop at 60 seconds.
6. The recorded pipeline uses separate transcription, text tutor/checker and speech-generation operations. It differs from a continuous live call.
7. Use the learner’s strongest supported language for maths. Do not spend most of the maths lesson correcting English unless English practice is the goal.

This guide explains the feature; reading it does not change any setting or grant access. Back to all help