Skip to main content
PromptQuorumBuilt for humans. Structured for AI.
Home/Power Local LLM
Voice, Speech & Multimodal

Weesper Neon Flow Review: Offline Dictation for Mac and Windows

·8 min readBy Hans Kuepper · Founder of PromptQuorum · Discovery engine for open-weight & open-source AI

Weesper Neon Flow is a closed-source dictation app for Mac and Windows that runs OpenAI's Whisper model on your own computer, and its maker says audio and dictated text never leave the machine. It costs 5 EUR a month, 45 EUR a year or 99 EUR once after a 15-day free trial. The newest release is version 1.8.12 from 28 October 2026. This review rests on the maker's public website, changelog and privacy policy and on its GitHub release page, not on hands-on testing.

Key Takeaways

  • Local claim: speech recognition runs on the device with Whisper; downloads, license checks, billing, updates and optional diagnostics still use the internet.
  • License and maker: proprietary, published by THE BLUE HOUSE (SASU) in Paris; the linked GitHub repository only hosts release installers.
  • Price: 15-day free trial, then 5 EUR a month, 45 EUR a year or 99 EUR once, with a 7-day money-back guarantee.
  • Maintenance: version 1.8.12 was released on 28 October 2026, so the project is actively updated.
  • Gaps: no Linux or mobile version, and the maker's own pages contradict each other on the free plan and on needing internet.

📍 In One Sentence

Weesper Neon Flow is a paid dictation app for Mac and Windows that, per its maker, runs Whisper on your own computer instead of a cloud service.

💬 In Plain Terms

Think of a microphone key that types for you: hold the shortcut, talk, let go, and your computer turns the speech into text without sending the audio to the maker.

Weesper Neon Flow is a paid, closed-source dictation app for Mac and Windows: you hold a shortcut, speak, and it types the cleaned-up text into whatever app has the cursor, running OpenAI's Whisper model on your own computer. This review covers what the maker's website, changelog and privacy policy document, where those pages disagree, and how it compares with other dictation tools in the directory.

Weesper Neon Flow Review: Offline Dictation for Mac and Windows

Get Weesper Neon Flow

Weesper Neon Flow installs like any desktop app, so no terminal is needed. Download the installer for your system from the official site; the speech models are fetched inside the app.

Platform
Get It
MacDownload for Mac — .dmg, macOS 13 or newer, Apple Silicon or Intel with AVX2
WindowsDownload for Windows — .msi, Windows 10 build 19041 or newer or Windows 11, x64 with AVX2
Free trial (15 days)Start the trial — all features, no credit card
Release notesChangelog and GitHub Releases
Websiteweesperneonflow.ai
Privacy policyWeesper Neon Flow privacy policy
Source codeClosed source; the GitHub repository hosts release installers only

The download page lists macOS and Windows only; no Linux, iOS or Android build is offered.

Weesper Neon Flow at a Glance

Attribute
Weesper Neon Flow
PlatformmacOS 13+ and Windows 10/11
Price15-day trial, then 5 EUR/month, 45 EUR/year or 99 EUR once
LicenseProprietary, closed source
RunsOn your computer with Whisper (whisper.cpp); internet for downloads and license checks
Languages50+ with automatic detection (the download page says 55+)
Speaker labelsYes, in Meeting mode (changelog 1.7.5)
Real-time transcriptionNot stated; text is inserted when you release the shortcut
Usable on CPU onlyNot stated; a modern multi-core CPU is the recommended baseline
Local API serverNot stated

Attributes follow the speech-to-text comparison used in the Local LLM Software Directory. A missing statement on the site means unknown, not unsupported.

What Weesper Neon Flow Is

Weesper Neon Flow is a push-to-talk dictation app, not a transcription editor. You hold a keyboard shortcut, speak, and release; the app runs Whisper on the computer and inserts cleaned-up text at the cursor, for example in Gmail, Slack, Notion or Word.

The maker is THE BLUE HOUSE (SASU), based in Paris, and the About page says the product was founded in 2024. It also sells a separate product, Weesper Transcribe, for audio and video files; this review covers only the dictation app.

This review draws on the website, changelog, privacy policy, terms and the GitHub release page. It involved no hands-on testing, so accuracy, speed and stability are not rated here.

How to Get Started

Setup is an install, a model download and a shortcut. The steps follow the website and its changelog.

  1. 1
    Install the app
    Why it matters: Download the .dmg for Mac or the .msi for Windows from the [download page](https://weesperneonflow.ai/en/download/) and start the 15-day trial; no credit card is needed.
  2. 2
    Pick a speech profile
    Why it matters: The default Small profile is 512 MB. The changelog says dictation works on a bundled model while the recommended one downloads.
  3. 3
    Hold, speak, release
    Why it matters: Click into any text field, hold the dictation shortcut, speak, and let go. A press-to-toggle mode was added in version 1.7.6.
  4. 4
    Tune the output
    Why it matters: Add words to the custom dictionary, set text shortcuts, and try the contextual and custom prompts the site lists.

Features

The feature set centres on fast, private dictation into any app. Everything below comes from the website and changelog.

  • Dictation anywhere. Text is inserted at the cursor in any app that accepts typing.
  • Clean-up. Automatic removal of filler words, plus smart punctuation and capitalization.
  • Languages. More than 50 languages with automatic detection, and speaking in one language to get text in another.
  • Meeting mode. Speaker diarization was added in version 1.7.5.
  • Personalization. A custom dictionary, text shortcuts, contextual prompts, custom prompts and a dictation history.
  • Microphone standby. Since version 1.7.10 the app no longer keeps the microphone open between dictations.

Models and Hardware

The app runs Whisper through whisper.cpp, with four speech profiles. The FAQ and changelog name Minimal (about 75 MB), Small (512 MB, the default), Medium (1.5 GB) and Large (3.2 GB), plus a Precision profile that was formerly called Turbo.

The download page lists a minimum of 4 GB RAM and 8 GB or more as recommended. On the Mac, an Apple Silicon M1 or newer with Metal is recommended; on Windows, a modern multi-core CPU or a supported Vulkan GPU. Both platforms need a processor with AVX2. The FAQ cites about 600 MB of disk space with the Small model. No speed or accuracy figures were reproduced here.

Privacy, License and Cost

The privacy policy says speech recognition runs locally and that audio, transcripts, dictated text and clipboard contents are not sent to Weesper servers for transcription. The policy was last updated on 23 July 2026. The terms add that the application operates 100% locally.

The internet is still involved elsewhere: downloads of the app and speech models, license activation and checks through a licensing service hosted on Cloudflare, billing through Stripe, update metadata from GitHub, and optional product analytics (PostHog) and crash diagnostics (Sentry). The policy says analytics events do not include audio, transcripts or dictated text. The code is closed and PromptQuorum did not examine network traffic, so these statements are the maker's own.

It is a paid, proprietary product: a 15-day free trial, then 5 EUR a month, 45 EUR a year or 99 EUR once, with a 7-day money-back guarantee. A personal license covers up to three of your own computers. The site does not say clearly whether prices include VAT, so check at checkout. The free plan after the trial is described two ways: 1,500 words a week in the FAQ and short clips of up to 60 seconds in the changelog.

Trade-Offs: Benefits vs. Limitations

Whisper on your computer

What it means in real use:
Dictation is processed on the machine, with no audio upload for transcription.
Limitation / caveat:
Closed source, so the claim cannot be checked; internet is still used for licensing and downloads.

Dictate into any app

What it means in real use:
One shortcut works in email, chat, notes and documents.
Limitation / caveat:
Mac and Windows only; no Linux, iPhone or Android version.

Lifetime option

What it means in real use:
A 99 EUR one-time purchase avoids a subscription.
Limitation / caveat:
Monthly and annual plans are the default; a free plan after the trial is described inconsistently.

Four speech profiles

What it means in real use:
Small and light models for older machines, larger ones for accuracy.
Limitation / caveat:
Large needs 3.2 GB of disk and more RAM; no speed figures are published.

Meeting mode with speakers

What it means in real use:
Speaker diarization is documented in the changelog.
Limitation / caveat:
Not tested here; the site gives no accuracy details.

Active development

What it means in real use:
Version 1.8.12 shipped on 28 October 2026.
Limitation / caveat:
Some pages, such as About, still carry older wording.

Weesper Neon Flow vs. Alternatives

These are the closest dictation tools in the directory: a free open-source app for three desktop systems, a Mac transcription app with dictation, a free Mac-only dictation app and a hybrid open-source app.

Weesper Neon Flow

Platforms:
Mac, Windows
Price and license:
Trial, then 5 EUR/month or 99 EUR once, proprietary
Role:
Push-to-talk dictation
Key difference:
Whisper on the device; paid and closed source

Handy

Platforms:
Mac, Windows, Linux
Price and license:
Free, MIT
Role:
Push-to-talk dictation
Key difference:
Open source and free; adds Linux
Articles about Handy (2)

Also mentioned in:

+9 more not shown

MacWhisper

Platforms:
macOS
Price and license:
Freemium, proprietary
Role:
Mac transcription app
Key difference:
Built for files and meetings first; Mac only
Articles about MacWhisper (2)

Also mentioned in:

FluidVoice

Platforms:
Mac
Price and license:
Free, GPL-3.0
Role:
Mac dictation app
Key difference:
Free and open source; Mac only
Articles about FluidVoice (1)

OpenWhispr

Platforms:
Mac, Windows, Linux, iOS
Price and license:
Freemium, MIT
Role:
Dictation with local or cloud modes
Key difference:
You choose local models, your own API key or its cloud
Articles about OpenWhispr (1)

Platform, price, and feature details for third-party apps change frequently. Verify current specifics on each project's own page before deciding.

Compare in directory

Who Should Use Weesper Neon Flow

  • Writers and office workers on Mac or Windows. One shortcut dictates into email, chat and documents.
  • People who avoid cloud dictation. Speech recognition stays on the computer, per the maker.
  • Multilingual users. More than 50 languages with automatic detection.
  • Buyers who prefer paying once. A 99 EUR lifetime license exists next to the subscriptions.

Who Should Not Use Weesper Neon Flow

  • Linux, iPhone and Android users. Only Mac and Windows builds are offered.
  • Anyone who needs auditable software. The code is closed and the privacy claims cannot be inspected.
  • Anyone who wants fully free software. Beyond the trial and a limited free plan, use is paid; Handy is free and open source.
  • Anyone who must work with no internet at all. Licensing and model downloads need a connection.

Frequently Asked Questions

Does Weesper Neon Flow work offline?

The maker says dictation works 100% offline once the app and a speech model are installed. The internet is needed to download them, to activate the trial or license and for occasional license checks. The About page says no connection is required, which conflicts with this. PromptQuorum did not test it offline.

Is my voice sent to the cloud?

The privacy policy says audio, transcripts, dictated text and clipboard contents are not sent to Weesper servers for transcription. The code is closed and PromptQuorum did not inspect network traffic.

How much does it cost?

After a 15-day free trial, 5 EUR a month, 45 EUR a year or 99 EUR once, with a 7-day money-back guarantee. Check at checkout whether VAT is included.

Which models does it use?

OpenAI Whisper through whisper.cpp, in four speech profiles from Minimal (about 75 MB) to Large (3.2 GB).

Who makes Weesper Neon Flow?

THE BLUE HOUSE (SASU), a company based in Paris, France.

Verdict

Weesper Neon Flow is a good fit if you want push-to-talk dictation on a Mac or Windows PC where, per the maker, Whisper runs on your own computer, and you accept paying for closed-source software. It has a free 15-day trial, a lifetime option and a recent release.

The caveats are scope and verifiability: no Linux or mobile version, closed code, privacy claims that PromptQuorum could not check, and maker pages that disagree on the free plan and on needing internet. Use the trial before you buy.

Sources

← Back to Power Local LLM

MCP