# SleekVoice vs Wispr Flow

> A sourced comparison of SleekVoice and Wispr Flow: where speech is transcribed, what it costs, which platforms exist, and what neither page should pretend to measure.

Canonical: https://sleekvoice.app/compare/wispr-flow
Updated: November 20, 2026

SleekVoice is a free, open-source voice typing app for macOS. Whisper runs on the Mac, including offline, and the text is typed into the focused app. Wispr Flow is a commercial dictation product from Wispr. The two are often compared because both aim at the same job: speak, and have text appear where you are writing. They do not make the same claims about where that speech is processed.

This page uses Wispr’s own pages for every statement about Wispr, and the SleekVoice site for every statement about SleekVoice. It does not include accuracy scores, speed tests, user counts, or ratings. None of those numbers are published here, so none are repeated from memory.

Checked against the public pages on 20 November 2026.

## Where is speech transcribed?

SleekVoice says transcription uses Whisper locally, including when the Mac is offline, and that your voice stays on your Mac.

Wispr’s [data-controls page](https://wisprflow.ai/data-controls) says: “Transcription always occurs on the cloud.” The same page says that is how Wispr chooses to provide transcription, and that turning the older “Privacy Mode” into a training switch does not change that. Wispr also says it keeps agreements so third-party AI providers do not train on customer data, and that you can separately allow or block Wispr’s own use of audio, transcript, and edits to improve Wispr’s models, in the app under Settings, Data and Privacy.

Those are different architectures. Local means the recording is processed on the computer you are speaking into. Cloud means the audio is sent to Wispr’s servers to be turned into text, which is what Wispr states. A switch that discards audio after transcription, or that opts out of training, is not the same thing as never uploading the audio. Wispr’s [help article on data sharing](https://docs.wisprflow.ai/articles/9609615338-private-cloud-sync-and-data-sharing-preferences-in-wispr-flow) says turning off “Improve the model for everyone” does not stop cloud processing.

## What happens to stored audio and history?

Wispr’s data-controls page says Dictation Cloud Storage controls whether Wispr stores transcripts, audio, and dictation history on Wispr’s servers, for features that need that history on more than one device. Their [privacy page](https://wisprflow.ai/privacy) says you can turn that storage off and that audio is discarded once it is transcribed. It also describes local history on the computer, with choices to keep it, delete it after 24 hours, or never save it, and it says mobile history, Notetaker, and cloud storage are managed separately.

SleekVoice does not publish a cloud history or a SleekVoice account. The personal dictionary and per-app modes are described as part of the local app. This page will not invent a retention timer SleekVoice has not written down.

## What does each one cost?

SleekVoice is free. The license is GPL-3.0. There is no paid tier on this site.

Wispr’s [pricing page](https://wisprflow.ai/pricing) says Flow is free on Mac, Windows, iPhone, and Android, and that plans add team management, admin controls, and support. The page’s own FAQ says Flow is free, includes dictation and the Notetaker with weekly limits, and that you do not need a credit card to start. Paid plans are also listed. Dollar amounts on that page depend on monthly versus annual billing and on which plan you are looking at, so this comparison does not freeze a price that the pricing page can change. Read the page for the current number.

Calling Wispr “a paid subscription only” would be incomplete. Calling the free plan “the same as SleekVoice” would also be incomplete, because Wispr still describes cloud transcription on every plan.

## Which platforms are supported?

SleekVoice is a Mac app. It does not publish Windows, iPhone, or Android builds.

Wispr’s pricing page says Flow Dictation is available on Mac, Windows, iPhone, and Android, and that the Notetaker runs on Mac, Windows, and iPhone, with Android described as later. If you need those platforms, SleekVoice is not the tool, and this page will not pretend otherwise.

## Can you read the source?

SleekVoice publishes its application source at [github.com/SuperAgents-Dev/SleekVoice](https://github.com/SuperAgents-Dev/SleekVoice) under GPL-3.0. You can read it, and pull requests are the way the project says it is built.

Wispr Flow is offered as a commercial app. This page does not claim Wispr has published or hidden its source. It only says SleekVoice’s source is public and Wispr’s website, on the pages reviewed here, presents a product you install rather than a repository you clone. If Wispr publishes source, their site is the place that would say so.

## What else is actually different?

| Question | SleekVoice | Wispr Flow |
| --- | --- | --- |
| Where speech is transcribed | On your Mac, including offline | Wispr says always in the cloud |
| Price | Free | Free plan and paid plans |
| License | GPL-3.0, source on GitHub | Commercial product |
| Platforms | macOS | Mac, Windows, iPhone, and Android for dictation, per Wispr’s pricing page |
| Dictionary | A personal dictionary on the Mac | Wispr documents a personal dictionary and, on paid team plans, a shared one |
| Account | No SleekVoice account on this site | A Wispr account is how their plans and data controls are described |

Wispr also documents Notetaker for meetings, context awareness that can read limited text from the focused app, and compliance features such as SSO on higher plans. SleekVoice has not published those features. They are not listed as missing scores. They are a different product surface.

Apple’s built-in Dictation is a third option, not an open-source one. Apple’s guide, [Dictate messages and documents on Mac](https://support.apple.com/guide/mac-help/use-dictation-mh40584/mac), says you can check Keyboard settings to see whether general text Dictation stays on device or is sent to Siri servers. That control is Apple’s. It does not describe SleekVoice or Wispr.

## What this page will not say

It will not say one product is more accurate. It will not quote a word-error rate. It will not say how many people use either app. It will not call Wispr careless, and it will not call local transcription magic. Local means the audio is processed on the Mac. The words you accept still land in the app you were writing in.

Security certifications are Wispr’s claims when they appear on Wispr’s pricing page, including SOC 2 Type II and ISO 27001. This site has not reviewed those reports. SleekVoice does not claim the same certifications.

## Questions

### Is SleekVoice an official Wispr product?

No. SleekVoice is a separate, GPL-3.0 project. Wispr does not publish it.

### Can I use SleekVoice without an account?

The project does not describe an account. Install the Mac app with `brew install --cask sleekvoice` or from GitHub releases, then grant the macOS permissions the system asks for. The steps are on the [getting started page](/docs).

### Where should I read Wispr’s current policy?

[Pricing](https://wisprflow.ai/pricing), [data controls](https://wisprflow.ai/data-controls), and [privacy](https://wisprflow.ai/privacy). Those pages can change. This comparison names the date it was checked.
