ReVerbal uses artificial intelligence technologies to support classroom accessibility through automated speech-to-text captioning and optional transcript summarisation.
How AI is Used
ReVerbal uses machine learning–based speech recognition by Deepgram to convert a teacher's spoken voice into live captions during classroom instruction. Where enabled, transcript summaries may be generated using OpenAI's GPT-4o-mini model.
AI functionality is limited to these purposes.
What We Do Not Do
ReVerbal does not:
- Build student or teacher profiles
- Monitor student behaviour or device activity
- Conduct emotion detection or biometric identification
- Store voice prints or create voice templates
- Use school data to train or improve AI models
- Make administrative or disciplinary decisions
Data Handling
Teacher audio is transmitted securely for real-time transcription and is not stored by ReVerbal.
Transcripts may be optionally saved by the teacher and are subject to defined retention limits.
AI models are configured so that customer data is not used for third-party model training.
Transparency
Captions and summaries generated by AI are clearly identified within the application.
ReVerbal publishes its sub-processor list and data processing locations. See our Privacy Policy, Data Processing Addendum, and AI Disclosure.
Material changes to AI processing or hosting arrangements will be communicated to customers in advance.
For further information, please refer to our Privacy Policy or contact support@reverbal.com.au.