ToText — Audio to Text

Eyal Ronen (Lovelace Loom)

ToText is an iPhone app from Lovelace Loom that transcribes recorded or imported audio in the cloud, then adds an AI summary with action items and lets you ask questions about the recording. The official pages state 60 transcription languages, a Free plan of 60 transcription minutes per month and Basic and Pro subscriptions bought through Apple. You can start without a ToText account, but processing needs an internet connection, and the pages warn that automatic transcripts and generated text can contain mistakes.

Why it fits

Audio to Text

iPhone / iPad

App Store
Good to know (3)

Check minute, upload and summary limits for your chosen plan and language.

Verify summaries and important details against the original document.

Free download does not mean every task is free. Check in-app purchases, export limits and country/device availability in the store.

Good to know
Good to knowiPhone / iPad
Free outputYes
No account neededYes
No watermark—
Works offlineNo
Batch processing—
Content languageTurkish

— Unverified conditions. Official documentation reviewed. The app itself has not been hands-on tested. Matches only documented content-language support, not interface language.

Sources and review date (6)
  • www.lovelaceloom.com/totext
  • www.lovelaceloom.com/totext/tr
  • apps.apple.com/us/app/totext/id6770080775
  • www.lovelaceloom.com/totext/legal/privacy-policy.html
  • www.lovelaceloom.com/totext/legal/terms.html
  • www.lovelaceloom.com/totext/support
Founder feedback and correction

Founder feedback and correction

If you represent this product, send a source-backed correction or update to the editors. Publication follows editorial review.

Editorial selection guide

Last checked:

Who is it for?

  • It suits people who want a phone-first recorder for meetings, lectures, interviews and voice notes: the product page says you can record through the device microphone or import a supported audio file, and the App Store description adds that you can share audio or video into the app from other apps.
  • It suits people who want more than a raw transcript: the product page describes a summary with action items, a transcript kept with the recording with speaker labels and word highlighting when available, tags, favorites, search, and questions whose answers cite a timestamp you can tap to replay.
  • It suits people who want to dictate into other apps: the product page describes a ToText keyboard that works in any app, and the support page says the Finish action inserts text without using ToText minutes, while Finish & Save keeps the recording and does use minutes.
  • It suits speakers of many languages: the pages state 60 transcription languages, including Turkish, English, Spanish, German, Japanese and Arabic, with automatic detection or a chosen primary language and an optional second one.

Getting started

  • Install ToText from the App Store on an iPhone with iOS 18.0 or later, the requirement shown in the listing. The listing says the Home, Insights, Calendar and Tasks screens need iOS 26. You can begin without a ToText account, and the Free plan needs no purchase.
  • Record a meeting, lecture, interview or idea with the microphone, or import a file: the pages list M4A, MP3, WAV, FLAC, OGG, WEBM, AMR and MP4, up to 4 hours within your remaining allowance. Choose the spoken language, or leave automatic detection on and optionally add a second language.
  • Read the summary and its action items, then check the transcript against the recording. The product page advises checking names, numbers and commitments against the recording, and says you can tap a cited timestamp to hear that part again.
  • Ask questions about the recording, keeping the per-recording question limits in mind (2 on Free, 30 on Basic, no limit on Pro under fair-use safeguards). Search saved transcripts, add tags and favorites, copy the text, save the summary as plain text, or export a PDF of the summary and transcript.
  • To dictate in other apps, open iOS Settings, General, Keyboard, Keyboards, Add New Keyboard, choose ToText and turn on Allow Full Access, which the privacy policy says dictation needs for the handoff to the app. Switch with the globe key and tap the Speak with ToText pill; the first tap of a session opens the app.

Decision points

  • Compare plans with how many minutes you record. The product page (United States prices, 11 October 2026) lists Free with 60 transcription minutes per month and up to 2 questions per recording, Basic with 1,000 minutes at $4.99 per month or $44.99 per year and up to 30 questions, and Pro with 2,000 minutes at $9.99 per month or $89.99 per year and no per-recording question limit, subject to fair use. The App Store listing shows the same four prices, and the Turkish page shows lira prices (Basic ₺199,99 per month or ₺1.799,99 per year; Pro ₺399,99 or ₺3.499,99), so check the price in your own store.
  • Know how minutes are counted. Allowances reset at the start of each calendar month in UTC and unused minutes do not roll over, and the support page says very short items are billed at a one-minute minimum. When the allowance runs out, existing recordings, search and export keep working and only new transcription waits. Paid subscriptions renew automatically unless cancelled at least 24 hours before the period ends, are managed in your Apple Account settings, and the terms say refunds are handled exclusively by Apple.
  • Decide whether the audio you record may leave your phone. The privacy policy (version 2.16, effective 17 October 2026) says audio is sent to a speech-to-text provider and the transcript to an AI provider, neither named, and that audio is deleted within about a minute of the job ending. It says the speech provider's published terms commit it not to retain audio for model training, that the AI provider does not use content for training and may retain it generally up to 30 days, and that data may be processed outside your country, including in the EU and the US.
  • Think about the optional ChatGPT connection. The pages say only text is shared, never audio. The policy says a separate copy of eligible recording text is kept on AWS in Frankfurt, Germany until you choose Settings, ChatGPT, Disconnect and delete, and that disconnecting only in ChatGPT, signing out or uninstalling does not delete it.
  • Check the keyboard's data path. The policy says the default Finish action uses Apple's speech recognition, trying Apple's on-device model first on iOS 26 and later, and that earlier versions or a failure may send audio to Apple's fallback service under Apple's policy, with no audio uploaded to ToText's servers. It also says the keyboard has no network access and does not log or transmit keystrokes.

Limits

  • The pages do not name the speech-to-text or AI providers and only describe them, so the underlying models cannot be identified from the official pages. The App Store privacy label is provided by the developer and not verified by Apple; it lists Purchases, User Content, Identifiers and Usage Data as linked to you, and User Content and Diagnostics as not linked.
  • ToText needs an internet connection: the product page says transcription, summaries and answers are processed in the cloud. The terms say transcripts and summaries may contain errors, omissions or hallucinations and are not professional advice, and the support page says accuracy varies with language, accent, audio quality and the number of speakers. Odia is supported for the interface and AI output but not for audio transcription.
  • Platform wording is thin. The vendor pages read do not name a device; the App Store lists iPhone with iOS 18.0 or later, iPad with iPadOS 18.0 or later, Mac with macOS 15.0 or later and an Apple M1 chip or later (marked Designed for iPad, not verified for macOS) and Apple Vision with visionOS 2.0 or later. No Android or web version is mentioned on the pages read.
  • Some numbers are missing from some pages. The terms mention a free tier with a monthly allowance without giving the number, and the FAQ refers to a remaining allowance without a figure; the 60-minute figure appears on the product page, the Turkish page and the App Store description. The terms also say features described as unlimited are subject to fair personal use, and prices differ by store and currency.
  • Consent and age rules are your responsibility. The terms forbid recording any person without the consent the law requires, say users under the age of majority may use the AI features only with a parent's or guardian's consent and say the optional ChatGPT connection is not for children under 13, while the App Store rating is 4+. The terms name Eyal Ronen as operator under the LovelaceLoom brand in Israel, and the about page gives no legal entity name.

Frequently asked questions

Is ToText free?

Partly. The product page and the App Store description state a Free plan with 60 transcription minutes each month and up to 2 questions per recording, with no purchase or ToText account needed. Basic (1,000 minutes) and Pro (2,000 minutes) are paid subscriptions through Apple, listed at $4.99 and $9.99 per month in the United States on 11 October 2026. Unused minutes do not roll over, and prices differ by store.

How do I transcribe an audio file with ToText?

Open ToText on an iPhone, import a supported file (M4A, MP3, WAV, FLAC, OGG, WEBM, AMR or MP4, up to 4 hours within your allowance) or record with the microphone, choose the spoken language or leave detection on, then read the transcript and the summary. The App Store description also says you can share audio or video into ToText from another app.

Does ToText work on Android or in a browser?

The pages read mention only the iPhone App Store listing. The vendor's product and support pages name no Android app and no web version, and the App Store lists iPhone, iPad, Mac (marked not verified for macOS) and Apple Vision. Check the vendor's site for any change.

Does ToText support Turkish?

Yes for transcription: the Turkish product page states that ToText supports 60 transcription languages including Turkish, and the English page lists Turkish among them. The App Store language list also includes Turkish. The support page says accuracy varies with language, accent, audio quality and the number of speakers, so check important details against the recording.

Does ToText work offline?

Not for transcription, summaries and answers: the product page says they are processed in the cloud and need an internet connection. The dictation keyboard's Finish action uses Apple's speech recognition; the privacy policy says iOS 26 and later try an on-device model first, while earlier versions or a failure may use Apple's fallback service, which may send audio to Apple.

What happens to my audio in ToText?

The privacy policy says audio is uploaded for transcription and deleted within about a minute of the job ending, with a storage rule sweeping leftovers about a day later, and that the finished transcript and summary wait on the server only until your phone downloads them, normally seconds. Your library lives on your devices and, when sync is enabled, in your personal iCloud; the policy says iCloud Drive sync is on by default and you can turn it off. Imported video may be sent in full when audio cannot be extracted on the device.

Official sources (8)
www.lovelaceloom.com/totext/www.lovelaceloom.com/totext/trapps.apple.com/us/app/totext/id6770080775www.lovelaceloom.com/totext/legal/privacy-policy.htmlwww.lovelaceloom.com/totext/legal/terms.htmlwww.lovelaceloom.com/totext/support/www.lovelaceloom.com/totext/faq/www.lovelaceloom.com/about/