Dictate in any app · Chat
Element users chose a decentralised, encrypted network for a reason worth respecting. With Typally you hold a hotkey, speak, and the text lands in Element (Matrix) exactly where your cursor was.
Disclosure: Typally is our product. We have tried to describe what Element (Matrix) already does without talking you out of using it.None. The composer is plain text.
Matrix is adopted by technical teams, privacy-conscious communities and organisations avoiding commercial platforms. That audience will ask what happens to dictated audio, and the answer needs to be direct.
In practice people use it for: Long technical messages, room topics, community moderation notes, and threaded discussion.
You do not configure anything in Element (Matrix). The hotkey works wherever your cursor happens to be:
Windows or macOS, then sign in. No card for the free tier.
Text goes wherever Element (Matrix) has focus. One click first saves undoing a paragraph in the wrong box.
Hold, talk, release. Long passages are fine.
Element's encryption protects the message in transit and at rest. It does not cover audio sent to a transcription service before the text exists.
Typally types into Element's composer like a keyboard, so long technical explanations and room announcements stop being compressed. Be clear-eyed about the trade-off: your message stays end-to-end encrypted, but the audio passes through a transcription service before it becomes text, so split what you dictate by sensitivity.
A technical community moderator writes the long explanation of a rule change by voice, having decided that public room announcements are not the sensitive category. The private discussions stay typed.
Split by sensitivity rather than trying to decide message by message. Most conversation is ordinary; the small remainder can be typed.
Element's encryption protects the message in transit and at rest. It does not cover audio sent to a transcription service before the text exists.
The message stays end-to-end encrypted. The audio, however, is processed by a transcription service on the way to becoming text.
It supports voice messages, which are audio rather than text in the composer.
Only fully local transcription, which is meaningfully less accurate today.
You can try this on the free plan — 100 minutes a month, no card, no trial countdown.
Also dictate in: Signal · Discord · Slack · LINE · Google Chat · see all apps