Speech-to-text software is a critical accessibility tool for some editors. While these tools recognise commands such as "period" or "backspace" to navigate text and handle punctuation, it seems likely that inputting things such as tables, headings, links and images would be more complicated without further support from tooling.

Let's investigate the VE experience for speech-to-text users to understand where we could better serve their needs.

• What's the experience like with iOS and Android accessibility tools?
• What's the experience with built-in tools for Windows, Mac and Linux?
• Expensive third-party accessibility tools?
• Which editing tasks are particularly frustrating?
  • Navigation within a page?
  • Opening/closing/using dialogs, especially for existing nodes?
  • Saving/cancelling edits
  • Creating links, images, tables?
  • Can you input "wikitext" shortcuts like == or {{, or does the IME interfere eg by adding spaces?
• Are we missing other opportunities to speed users up?
• Who are our speech-to-text users?
  • Users who prefer using speech
  • Users with access needs
  • Users who are also using TTS, eg, blind users on a phone
  • Do they use VE or raw wikitext right now?
• How do we investigate?
  • Are there blogs / village pump posts / discord discussions about VE?
  • Podcasts, videos, blogs about the experience in general?
  • We should talk to users with access needs directly wherever possible