Transcribe audio to text on your iPhone
Record a meeting, a lecture or a voice memo, or import a file you already have. VoText turns the voice into text in 33 languages, labels who is speaking and timestamps each line.
Speech to text at a glance
- What it transcribes
- Recordings made in VoText, and audio or video files from Files, Voice Memos and other apps
- Languages
- 33, detected automatically, even when speakers switch languages mid-sentence
- Speakers
- Each speaker labeled, and you can give them names
- Timestamps
- On every line; a tap plays that moment of the recording
- Longest file
- 2 hours, or 5 hours with Pro
- Export
- PDF, plain text, Markdown, SRT and WebVTT subtitles, and the original audio
- Free option
- On-device transcription, offline and without credits, in the languages your device recognizes
- Privacy
- Audio is deleted from our servers within 24 hours, and the AI services may not train on it
- Price
- 1 credit per minute of audio. 20 free credits to start, then from about $0.27 per hour
What people transcribe
- Voice memos and notes
- Talk an idea through on a walk and read it back as text. Share a recording from Voice Memos to VoText, or import it from Files.
- Meetings and interviews
- Speaker labels show who said what, and a tap on any line plays it again, so a quote is easy to check.
- Lectures and classes
- Record the whole lecture, then search the transcript for what was said instead of scrubbing through audio.
- Videos and podcasts
- Import a video and VoText transcribes its sound. Export SRT or WebVTT subtitles, timed to the audio, for your editor or player.
How to transcribe audio on iPhone
Record or import
Tap record, or import an audio or video file. Recording is free, and the audio stays on your device.
Choose how to transcribe
In the cloud, with speaker labels and language detection, or on your device, free and offline. VoText shows the cost before anything starts.
Read and correct
Tap a word to fix it, name the speakers, and tap a timestamp to hear that moment again.
Export or share
Save a PDF, plain text, Markdown or subtitles, or share the original audio.
Speakers, timestamps and languages
- Who said what
- Cloud transcription tells voices apart and labels each speaker. Give them names right in the transcript.
- A timestamp on every line
- Each line carries the moment it was said. Tap it to play that moment and check a word against the recording.
- The language, recognized for you
- There is no language to pick first. When a conversation switches language halfway through, each part is written in the language it was spoken in.
Cloud or on-device transcription
Two ways to turn speech into text. Use whichever suits the recording.
| Feature | Cloud | On device |
|---|---|---|
| Cost | 1 credit per minute, billed by the second | Free |
| Internet connection | Needed | Not needed |
| Speaker labels | Yes | No |
| Languages | 33, detected automatically, also mixed in one recording | One at a time, among those your iPhone or iPad recognizes |
| Where the audio goes | To trusted AI services; deleted from our servers within 24 hours | Nowhere. It never leaves your device |
Export your transcript
- A clean document to read, print or send.
- Plain text
- Paste it into an email, a note or any app.
- Markdown
- For note-taking apps and documents that read Markdown.
- SRT and WebVTT
- Subtitles timed to the audio, for video editors, players and the web.
- Original audio
- The recording itself, to keep or share alongside the text.
Your recordings stay yours
- Recordings and transcripts are stored on your iPhone or iPad.
- Audio leaves your device only when you ask for a cloud transcript, and only after you agree.
- Our servers delete it within 24 hours, and the AI services that transcribe it may not use it to train their models.
- No account needed, no ads and no tracking.
What it costs
One credit transcribes one minute of audio. New users get 20 free credits. Plans start at $6.99 a month, and an hour of transcription costs from about $0.27.
Prices in US dollars. The App Store shows the local price for your country, including taxes, before you buy.
Speech to text questions
Can VoText transcribe recordings from Voice Memos?
Yes. In Voice Memos, share the recording and choose VoText, or save it to Files and import it in VoText.
Can I transcribe a video?
Yes. Import a video file and VoText transcribes its sound. You can then export the transcript as SRT or WebVTT subtitles.
Is there a free way to transcribe audio?
Yes. On-device transcription is free and works offline, in the languages your iPhone or iPad recognizes. New users also get 20 free credits, enough for 20 minutes of cloud transcription.
How long can a recording be?
Cloud transcription takes files up to 2 hours long, or up to 5 hours with Pro.
Which languages can VoText transcribe?
33 languages. VoText recognizes the language itself, also when people switch mid-sentence. See the full list.
How do I get the most accurate transcript?
Record close to the people speaking and away from noise. For a recording in one language, choosing that language before you transcribe helps with names and technical terms. Any word can be corrected afterwards.
Is my audio used to train AI?
No. The AI services VoText uses may not train their models on your audio, and our servers delete it within 24 hours. See the services we use.