Hotline
Hotline
One tap. Your AI is listening. Press once, on your wrist or from a shortcut, and talk: the transcript is delivered to your own assistant, your notes, or any tool you connect.
iPhone and Apple Watch. No accounts, no data collected. Coming to the App Store.
Press once, from anywhere
There is no app to open and no button to hunt for. Start from the Watch on your wrist, a shortcut, the Action button, Control Center, the Home Screen widget or the Lock Screen, and talk. A Live Activity keeps the recording and its stop control on the Lock Screen while it runs.
The best models, preconfigured, or Apple’s on device
Groq Whisper, OpenAI GPT-Transcribe and ElevenLabs Scribe v2 are already set up, and your own OpenAI-compatible server works too. Paste that provider’s key into Settings and the audio goes straight from your device to the provider. There is no proxy and no subscription from us.
Or leave it on Apple’s on-device model, which is the default. It is free, it works offline, it needs no account and no API key, and the audio never leaves the phone. A fresh install can record and get a transcript with nothing configured.
Sent anywhere your assistant runs
Point Hotline at any HTTPS URL and each finished transcript arrives there as JSON, signed with HMAC-SHA256 so you can verify it came from your device. That is how it reaches your own assistant, your notes, or any tool you connect. Failed deliveries retry on their own, and the queue survives a restart. Plain HTTP is accepted on localhost, .local hosts and private networks, so an agent on your own Wi-Fi works without a certificate.
Meetings, with who said what
Mark a recording as a meeting while it runs and the provider returns speaker labels. The transcript comes back as labelled paragraphs and the webhook body carries the same segments with their speakers and timings. Weeks later it is still there to search and read back.
One press from your wrist
The Apple Watch app runs on its own. It hands the audio to your iPhone when the phone is in reach. With a cloud model and your key, it transcribes and delivers from the wrist when the phone is not; with the on-device model it keeps the recording and hands it off as soon as the phone is back. Either way the recording lands in your phone’s history, and the webhook is sent exactly once.
Playback that follows the transcript
Play a recording back and the transcript highlights along with the audio. Tap a word to seek to it.
Private by architecture
No accounts. No login. No servers of ours. No analytics. Your provider keys stay in the device keychain, audio goes only to the transcription provider you chose, and transcripts go only to the webhook you configured.
More
- Webhook reference: the payload, the signature, retries and a working receiver.
- Support: how it works, what to check first, and how to reach us.
- Privacy: what the app does with your data, in plain sentences.