Features · Voice
Hold it. Say it. Let go.
A big microphone, a short message, and your words appear as text before the helper starts. No keyboard needed.
The microphone is the front door.
Hold the button to talk and let go to send, or tap to lock it open when your hands are busy. Your voice is written out, shown to you as text, and the helper answers it like any other message. If the microphone is blocked or missing, the page says so and you can type instead.
- Hold to talk, tap to lock, release to send
- A message can be up to 2 minutes 10 seconds
- Your words show as text, marked as voice
What happens to your voice
Four steps. Then the audio is gone.
You record
The browser records the message. It needs your permission once.
It is sent to be written out
A speech service turns the audio into text. The workspace first tries Groq (Whisper large-v3-turbo) and falls back to OpenAI (gpt-4o-mini-transcribe).
Your names help
The names in your workspace, such as clients and projects, go along as a hint so "Tjerk" does not become "Jerk".
Only the text is kept
The audio file is not kept after it was read. What stays is the text of your message.
Honest labels
What to expect.
- Pre-betaIt depends on the provider keys
Speech writing runs through the providers named above once their keys are set on your workspace. Without a key the app uses a labelled stand-in transcript, so you always see what is real.
- LiveLimits that keep the bill honest
One message is at most 6 MB and 2 minutes 10 seconds, and a workspace can send up to 150 voice messages a day.
- PlannedLive dictation and a spoken reply
Words appearing while you speak, on-device writing-out on the phone, and the helper answering out loud. Planned, no date.
- Not offeredNo recordings stored
The app is set not to keep the audio after it is read.
Questions
About talking.
How does the voice part work?
Hold the microphone and talk, or tap to lock it open. The recording is written out as text by a speech service, shown to you, and the helper answers it like any other message. See voice messages.
Is my recording kept?
The app is set not to keep the audio after it has been read. What stays is the text of your message, like a typed one.
How long can a message be?
Up to 2 minutes 10 seconds and 6 MB. A workspace can send up to 150 voice messages a day.
Who hears it?
Your recording goes to a speech service to be written out: by default Groq (Whisper large-v3-turbo), with OpenAI (gpt-4o-mini-transcribe) as the fallback. The names in your workspace go along as a spelling hint. See the privacy statement.
Can it listen while I talk and answer out loud?
Not yet. Live dictation and a spoken reply are planned, with no date.