Stillword for iPhone
Transcription that stays on your iPhone.
Turn audio and video into editable text, privately. Stillword runs Whisper speech recognition on the device, so your recordings are never uploaded.
Stillword isn’t on the App Store yet. This page will link to it when it is.

At a glance
How it works
From a recording to text in three steps.
Step 1
Choose a file
Pick a video from Photos, or an audio or video file from Files.
Step 2
It transcribes on your iPhone
Stillword extracts the audio and runs Whisper locally. Keep the app open while it works.
Step 3
Edit, search, export
Fix any line, find what was said, then share a TXT, SRT or VTT file.
Features
Everything the app actually does.
Transcribes on your iPhone
Speech recognition runs on the device with Whisper. Your audio and video are never uploaded to Stillword or anyone else.
Works offline
Download a model once. After that, transcription works without a connection, on every plan.
Audio and video
Import from Photos or Files. Stillword reads the sound from a video and transcribes that.
Three Whisper models
Choose Tiny for speed, Base for balance or Small for the best accuracy, depending on your plan.
Editable transcripts
Correct any line. Each line keeps its time, and you can revert to the original transcription.
Timestamps
Transcripts are stored in timed segments, so you can see where every line starts.
Library and search
Every transcript is saved on your iPhone. Search the Library, rename items and find words inside a transcript.
TXT, SRT and VTT export
Share the text as a TXT file, or subtitles as SRT or VTT with Plus or Pro. Export all your transcripts as a backup.
Light, Dark and OLED
A quiet, monochrome design with three themes, including true black for OLED screens.
Speech models
Choose speed or accuracy.
Models are Whisper speech models, downloaded once from Hugging Face. They are stored on your iPhone and stay there until you delete them.
| Model | In the app | Download size | Included in |
|---|---|---|---|
| Whisper Tiny | Fast | about 78 MB | Free and above |
| Whisper Base | Balanced | about 148 MB | Plus and above |
| Whisper Small | High accuracy | about 488 MB | Pro |
Larger models are more accurate and slower, and use more memory. Accuracy depends on the recording: clear speech, little noise and one language at a time work best.
Formats
Audio and video in.
Audio
Video
Very long files use a lot of memory on the larger models. If one doesn’t finish, try a smaller model or a shorter file.
Privacy
What stays on your iPhone, and what doesn’t.
Stays on your iPhone
Your audio, your video, the text and the transcripts you save. Stillword has no account system and no server that receives your recordings.
Uses the internet only for
Downloading a speech model, and handling your subscription through Apple and RevenueCat. Stillword has no analytics or crash-reporting service.
Pricing
Start free. Upgrade when you need more.
Free
A$0
30 minutes of transcription per month. Whisper Tiny.
Plus
A$5.99/month or A$44.99/year
180 minutes of transcription per month. Whisper Tiny and Base.
Pro
A$7.99/month or A$59.99/year
Unlimited transcription. Whisper Tiny, Base and Small.
Get Stillword
Coming to the App Store.
Stillword is built for iPhone. Subscriptions are bought and managed through Apple’s App Store, never on this website.