Best Speech to Text Apps for iPhone and iPad in 2026: On-Device vs Cloud

The short version:

• Best overall for private transcription: LoroNote — Whisper Large V3 Turbo runs on the device, with speaker labels and TXT/SRT export
• Best free starting point: Apple Voice Memos — already installed, good enough to find out whether you need more
• Best for team meetings: Otter — real-time cloud transcription with Zoom, Google Meet, and Teams integration
• Best one-time purchase: Aiko — upfront price, on-device Whisper, and it also runs on Mac
• Best when a human transcript is required: Rev — AI drafts plus 99%-accurate human transcription as a paid option

What this list covers, and what it does not

There are two different jobs people mean by “speech to text” on an iPhone, and mixing them produces a useless list.

The first is transcribing a recording: you have a meeting, lecture, interview, or voice memo, and you want text out of it. The second is dictation, where you speak into a text field instead of typing, and the words appear in Messages, Mail, or Slack as you go. Apps like Wispr Flow sit firmly in that second category and compete with the built-in iOS keyboard microphone, not with the apps below.

This list is about the first job. Every app here takes audio — recorded in the app or imported — and gives you a transcript you can edit, search, and export.

The single biggest split within that group is where your audio goes. Some apps run speech recognition on the iPhone itself; others upload the file to a server. That difference decides what happens to a confidential client call, whether the app works on a plane, and what you are agreeing to when you tap accept. We put it in the first column of the comparison for that reason.

This list was compiled by the LoroNote team on August 13, 2026. LoroNote is our own app, so we have stated our criteria before the rankings, described every other app using its own US App Store listing, and marked anything the listing does not say as not stated rather than guessing.

What to look for in an iOS speech to text app

• Where audio is processed. On-device means the recording never leaves your iPhone and transcription works in Airplane Mode. Cloud means a server does the work, which usually buys you real-time collaboration and heavier AI features. Neither is automatically right, but you should know which one you picked. Our offline transcription guide explains the data flow in detail, and our offline-only shortlist ranks apps by that criterion alone.
• Language coverage. Whisper-based apps typically list around 100 languages. Apps built on other engines often support far fewer. If you record in anything other than English, check the listing before paying. Which Whisper variant an app actually runs matters too — we broke that down in the best Whisper apps for iPhone.
• Speaker separation. For a two-person interview you can usually cope without it. For a six-person meeting, a wall of unattributed text is close to worthless. See how speaker diarization works.
• What you can get out. Plain text, timestamped text, and SRT subtitles are different deliverables. Some apps deliberately do not let you edit inside the app at all.
• How it charges. Free with in-app purchases, one-time purchase, and per-minute quotas behave very differently once you have fifty recordings.
• Device requirements. On-device models need recent hardware. Several apps in this list require iOS 18 or later, and one requires iOS 26.

The apps compared at a glance

Everything below was read from each app’s US App Store listing on August 13, 2026. Ratings are US storefront figures; the US store is only one of roughly 170, so a small US count does not mean a small worldwide count.

LoroNote’s worldwide figure across all storefronts is 4.7 from 279 ratings. We list the US number in the table so every row uses the same basis.

1. LoroNote — best overall for private transcription

LoroNote runs OpenAI’s Whisper Large V3 Turbo directly on the iPhone or iPad. Audio and text are processed locally, so a confidential call becomes a transcript without a recording ever reaching a transcription server, and the whole thing works in Airplane Mode once the app and its resources are installed.

The recording and the transcript stay connected

Most of the work with a transcript happens after transcription finishes. LoroNote keeps the text tied to the audio: highlighted words follow playback, and tapping a line jumps to that moment in the recording. That turns checking a misheard number into a two-second job instead of scrubbing a waveform.

Around that sit the parts that matter at volume rather than on your first file — folders, a calendar view, search, batch find-and-replace across notes, and a custom dictionary for names, acronyms, and technical terms that generic models reliably mangle.

Speakers are separated on the device

Speaker identification also runs locally. You can rename speakers, and the labels survive when you share the transcript. On supported devices, Apple Intelligence can then summarize the transcript or pull out action items without sending it anywhere either.

Capture is broader than most apps here: record in the app, record from Apple Watch, or import MP3, WAV, M4A, MP4, and MOV files. Exports are TXT for editable text and SRT for subtitles. For the reasoning behind the model choice, see why LoroNote uses Whisper Large V3 Turbo.

Best for: meetings, interviews, lectures, and voice memos you would rather not upload, especially in languages beyond English.

Keep in mind: on-device transcription needs recent hardware — iPhone 13 or later, iPhone SE (3rd generation), or a supported recent iPad, on iOS 18.6 or later. There is no Mac, Windows, web, or Android version, and long recordings take real time and battery because the work happens on your device rather than on a server.

2. Apple Voice Memos — the free baseline you already have

Voice Memos ships with iOS and holds a 4.8 rating from over a million US ratings. It records, syncs through iCloud across iPhone, iPad, Apple Watch, Mac, and Vision Pro, and includes editing tools like trim, replace, and Save As. Apple’s description confirms a transcription feature that “converts speech into text, simplifying the process of finding and referencing specific content.”

What the App Store description does not say is which languages transcription supports or where it runs. Both depend on your device and iOS version rather than on the app itself, so the listing is genuinely silent rather than evasive.

Start here. If the transcript quality, language, and export options are enough, you have saved yourself a purchase. Our guide on turning iPhone recordings into text walks through exactly what the built-in tools do and where they stop.

Best for: finding out whether you need a dedicated app at all.

Keep in mind: no speaker labels, no SRT export, and no imported-file workflow. It is a recorder with transcription attached, not a transcription tool.

3. Otter — best for team meetings

Otter is the most established cloud option on iOS, with a 4.8 rating from over 72,000 US ratings. It transcribes in real time, joins Zoom, Google Meet, and Microsoft Teams meetings, captures slides automatically, and generates summaries. Notes are searchable, shareable, and editable by teammates, which is the part a local app fundamentally cannot replicate.

The listing states availability in English, Spanish, French, and Japanese. That is a much narrower set than the Whisper-based apps here, and it is the deciding factor if your recordings are in another language.

Your audio goes to Otter’s servers. For a public webinar that is unremarkable; for a privileged legal call it is a decision worth making deliberately. We compared the two models directly in LoroNote vs Otter.ai.

Best for: recurring team meetings where several people need the same live notes.

Keep in mind: four languages, cloud processing, and collaboration features that only pay off if your colleagues also use it.

4. Aiko — best one-time purchase

Aiko costs $24.00 upfront with no subscription, and runs Whisper locally. Its listing is unusually direct: “Nothing leaves your device,” and it “favors accuracy over speed.” It supports audio in 100 languages, exports subtitles, and includes automatic word replacement for recurring mistakes.

It is also candid about what it does not do. The listing states that Aiko does not do live transcription while recording, does not currently have speaker detection, and — per its FAQ — is not planning in-app text editing; the intended workflow is to export and edit elsewhere.

Unlike everything else in this list, Aiko runs on Mac and Apple Vision as well. It now requires iOS 26 or later, which is a real constraint on older hardware. We looked at it in more depth in LoroNote vs Aiko.

Best for: importing finished recordings, getting accurate text, and moving it into another app — across iPhone, iPad, and Mac.

Keep in mind: iOS 26 minimum, no speaker detection, no live transcription, no in-app editing.

5. Just Press Record — best for capture speed

Just Press Record is a $6.99 one-time purchase built around getting a recording started fast: Lock Screen, Control Centre, the Action Button, Shortcuts, Siri, a widget, or independently on Apple Watch with a later sync. Recording time is unlimited and it syncs across iPhone, iPad, and Mac through iCloud.

Transcription supports over 30 languages, independent of your device’s language setting, with synchronized text highlighting during playback and punctuation commands as you speak. You can edit both the audio and the transcript in the app.

Its listing does not state where transcription runs, so we have not classified it either way. At a 4.3 rating from about 1,500 US ratings, it is the lowest-rated app in this list, though that is still a solid score.

Best for: capturing ideas the instant they happen, especially from an Apple Watch.

Keep in mind: 30+ languages is well short of the Whisper-based apps, and no speaker labels.

6. Transcribe – Speech to Text — best cloud all-rounder

Transcribe handles both live speech and uploaded audio or video, in over 100 languages and dialects, and translates finished transcripts into more than 40. Its listing specifically mentions timestamps and speaker separation, plus AI summaries and rich-text formatting. It runs on iPhone, iPad, and Mac, and includes 30 free minutes to try.

This is the broadest feature set per dollar in the list, and the trade-off is straightforward: it is a cloud service, so your files are uploaded.

Best for: multilingual work where translation matters as much as transcription.

Keep in mind: cloud processing, and free minutes run out quickly — check the current in-app pricing before committing a large backlog.

7. Rev — best when accuracy has to be guaranteed

Rev is the only app here that offers human transcription. You get live AI transcripts in the app, and can order human transcripts advertised at 99%+ accuracy when a draft is not good enough. It is aimed at legal, research, and media professionals, with team collaboration, AI templates, and the ability to query your transcripts.

Its listing notes that call recording is not supported, which matters if that was your plan.

Use Rev when a transcript is going into a court filing or a published article and machine accuracy is not sufficient. For everyday notes, the paid human tier is more process and cost than the job needs.

Best for: deposition, interview, and archival transcripts where errors carry real consequences.

Keep in mind: human transcription is a paid service with turnaround time, and everything is processed in the cloud.

How to choose for your own recordings

• If the audio is confidential — client calls, patient conversations, unpublished research — start with an on-device app and rule it out only if something is genuinely missing.
• If you need live shared notes with colleagues, a cloud service is the only real answer. Pick one whose language list actually covers your recordings.
• If you record in anything other than English, Spanish, French, or Japanese, filter by language coverage first. It eliminates more options than any other criterion.
• If you have a backlog of existing files, check import formats before anything else. Several apps only transcribe what you record inside them.
• If you are not sure yet, open Voice Memos and transcribe something today. It costs nothing and tells you what you actually need.

Whichever you pick, test with three to five minutes recorded in your normal environment — your microphone, your room, your accent, a few names and numbers and technical terms. Then check what is missing, how the punctuation reads, whether speaker changes are marked, and how long it took. A list ranks apps in general; your own recording tells you whether the result is usable.

Frequently asked questions

What is the best speech to text app for iPhone?

It depends on which job you are hiring it for. For private transcription of your own recordings, LoroNote — on-device Whisper Large V3 Turbo with speaker labels and SRT export. For live team meeting notes, Otter. For a one-time purchase that also runs on Mac, Aiko. For guaranteed accuracy on high-stakes transcripts, Rev’s human tier. And before paying for anything, Voice Memos is free and tells you whether the built-in transcript is already enough.

Which iOS speech to text app works without an internet connection?

LoroNote and Aiko both state that speech recognition runs on the device, so transcription does not require a network connection once the app and its resources are installed. Cloud services such as Otter, Transcribe, and Rev need connectivity to produce a transcript.

Is the free iPhone transcription good enough?

For short personal memos in a well-supported language, often yes. Voice Memos is free, already installed, and syncs everywhere. It becomes limiting when you need speaker labels, SRT subtitles, imported files, or a language it does not handle well.

Which app is best for transcribing meetings with several speakers?

If the meeting can go to a cloud service, Otter is built for exactly that and adds live shared notes. If it cannot, LoroNote performs speaker identification on the device and lets you rename speakers before sharing the transcript.

Do any of these apps transcribe imported audio and video files?

LoroNote imports MP3, WAV, M4A, MP4, and MOV. Aiko and Transcribe also work from imported files. Voice Memos is built around its own recordings rather than an import workflow.

How accurate is on-device transcription compared to cloud transcription?

Close enough that the choice usually comes down to other factors. On-device Whisper models are the same family used by many cloud services; the practical differences are speed on your particular hardware, and whether the service adds post-processing. Accuracy on your own audio — accent, microphone, background noise, vocabulary — varies far more than the gap between engines, which is why a five-minute test is worth more than any published benchmark.

Final thoughts on the best speech to text apps for iOS

There is no single best speech to text app for iPhone, because the apps are not solving the same problem. Otter is built for teams that meet constantly. Rev exists because sometimes a machine transcript is not admissible. Voice Memos is free and already on your phone. Aiko is a focused utility with an honest list of things it will not do.

What has changed in 2026 is that the private option is no longer the compromise. Running Whisper Large V3 Turbo on an iPhone gives you the same multilingual model family the cloud services use, with speaker labels, synchronized playback, and subtitle export — and the recording never leaves your device.

If your audio is worth protecting, start there and see whether you miss anything.