Support

Does FerIQ work offline?

Yes. After the one-time download of the Whisper model (~600 MB) on first launch — and the FluidAudio speaker-diarization models (~100 MB) if you enable speaker labels — the app works entirely offline. You can record and transcribe without an internet connection.

Why is the first launch slow?

On first launch FerIQ downloads the Whisper large-v3-turbo model. Depending on your internet speed this can take 1–5 minutes. The download only happens once; subsequent launches are instant.

Why does FerIQ need microphone permission?

So it can record audio for transcription. The microphone is only active while you are explicitly recording. We do not stream microphone audio anywhere.

Are my recordings shared with anyone?

No. Your recordings stay on your device. If you use iCloud, they may be synchronised between your own devices through your private iCloud database — but never to our servers. We have no way to access your data. See our Privacy Policy for full details.

Which languages does FerIQ support?

The transcription picker lists 14 languages: Danish, English (US), German, French, Spanish, Italian, Dutch, Brazilian Portuguese, Swedish, Norwegian, Japanese, Simplified Chinese, Korean, and Polish. Setting it to Auto lets Whisper detect and transcribe any of its 99 supported languages per utterance — including languages not in the picker like Arabic, Hindi, Russian, Turkish and more.

Translation supports the same 14 target languages.

The app's interface (menus, buttons, settings) is localised to 11 languages: English, Danish, German, French, Spanish, Norwegian, Swedish, Japanese, Simplified Chinese, Korean, and Polish.

What features require Apple Intelligence?

AI summary, action items extraction, chapter detection, and translation use Apple's on-device Foundation Models. These require iOS 26 / iPadOS 26 / macOS 26 on supported hardware (iPhone 15 Pro and later, M-series iPads, M-series Macs). If Apple Intelligence is unavailable on your device, FerIQ still works for transcription, recording, speaker labels, and editing — only the AI-powered features are gracefully disabled.

How accurate are the speaker labels?

FerIQ uses FluidAudio's offline diarization pipeline to label each transcript segment as "Speaker 1", "Speaker 2", and so on. You can then rename any speaker — the new name propagates across the entire transcript. Accuracy is best on recordings with two or three distinct speakers, clear audio, and minimal overlap. The first time you use this feature, FerIQ downloads ~100 MB of models in the background.

How does the AI summary work?

After your transcript is ready, tap the sparkles button to generate a summary. The transcript is sent to Apple's on-device Foundation Models for a quick or detailed summary depending on the style you pick. For longer transcripts (over ~9,000 characters), the app automatically splits the content into chunks and combines the partial summaries — so even very long meetings produce a coherent overview.

What's the difference between Quick Mode and the full file pass?

By default FerIQ runs a refinement pass after you tap Stop, which can take several minutes on long recordings. Quick Mode skips this and uses the live preview segments as the final transcript — instant, with about 2% lower accuracy. For recordings over 15 minutes, Quick Mode is automatically enabled unless you explicitly turn it off in Settings.

How do I export transcripts?

Open any recording and tap the share icon to export as plain text, Microsoft Word (.docx), or PDF. You can also bulk-select multiple recordings from the Library and export them all as a zipped bundle including the audio files.

What are meeting templates?

Premium users can pick from eight purpose-built templates — interview, sales call, lecture, standup, decision meeting, brainstorm, 1:1, or generic — that shape the AI summary and action items to the kind of recording. Pick a template before you record or apply one to an existing recording.

How do I import an existing audio file?

Tap the import button on the Library view, drag-drop an audio file onto the app window (Mac), or use Files / Share Sheet (iPhone / iPad). Supported formats include CAF, WAV, M4A, MP3, AAC, FLAC, and OGG.

How do I delete recordings?

Swipe a recording in the Library to delete one at a time, or use Select → Delete Selected for multiple recordings. To permanently remove every recording on the device, go to Settings → Privacy → Delete all my recordings. Deletions sync to your other iCloud devices.

What's the keyboard shortcut to start recording on Mac?

⌘⇧R from anywhere in macOS starts a new FerIQ recording (the global shortcut is registered system-wide). ⌘N inside the app opens the recording sheet.

Can I use Siri or Shortcuts?

Yes. Two App Intents are exposed: "Start a new FerIQ recording" and "Open my latest FerIQ recording" — both work in Siri, Shortcuts, and on the Lock Screen / Action Button. There's also a "Search transcripts" intent for searching across your saved transcripts.

How does pricing work?

FerIQ has a free tier with full recording and on-device transcription. Premium (29 kr/month, 249 kr/year with a 30-day free trial, or a one-time Lifetime purchase at 599 kr) unlocks AI summary, action items, speaker labels, meeting templates, and advanced exports. Pricing varies by region as set by the App Store. Manage subscriptions at App Store → Account → Subscriptions.

Can I get a refund?

Refunds are processed by Apple, not by the developer. Request one at reportaproblem.apple.com.

Why does a translation sometimes get rejected?

FerIQ verifies that Apple Foundation Models actually produced text in the language you asked for. If the result comes back in the wrong language (the model occasionally returns the source unchanged, or refuses on safety-classifier edges), the segment is marked as failed and shown with an orange "Translation rejected — retry?" banner. Tap retry, switch to a different target language, or report the segment if it persists.

Something else?

Email fyigen@gmail.com. We read every message.