Spokenly and Wispr Flow both put your words wherever your cursor is, on a computer or a phone, but they're built on opposite ideas. Spokenly gives you choices: free local models that run on your own device, Spokenly's cloud models on Pro, or cloud providers you pay directly with your own API keys. Wispr Flow makes the choices for you: everything runs in its cloud, with cleanup, formatting and a meeting Notetaker built in.
Here's how they compare as of October 20, 2026, based on each company's website, docs, policies and store listings. We make a third app, LumeVoice, and we haven't used either of these ourselves, so this compares what each vendor documents rather than reporting a hands-on test.
The short verdict#
Choose Spokenly if you:
- want unlimited dictation for free, with local Whisper and Parakeet models
- need to dictate offline, or must keep audio on your device
- use Linux or an iPad, or want a Chrome extension
- like plugging in your own API keys from OpenAI, Deepgram, Groq and others
- want to transcribe audio and video files
- want the cheaper Pro plan: $99.99 a year against $144
Choose Wispr Flow if you:
- use an Android phone
- want cleanup and formatting that work out of the box, with no models or instructions to set up
- want AI meeting notes on Windows as well as Mac
- need security paperwork for work: SOC 2 Type II, ISO 27001, single sign-on or a signed BAA
- want a free plan that includes cloud transcription
Spokenly vs Wispr Flow at a glance#
| Spokenly | Wispr Flow | |
|---|---|---|
| Where speech is processed | Your choice: local models, Spokenly's cloud (Pro) or your own provider keys | Always in Wispr's cloud |
| Works offline | Yes, with local models | No |
| Desktop | Mac (macOS 13.3+), Windows 10 and 11, Linux | Mac (macOS 12+), Windows 10 and 11 (x64 only) |
| Phone and tablet | iPhone and iPad (iOS 16.4+); no Android | iPhone (iOS 18.3+, no iPad), Android 13+ |
| Free plan | Unlimited local dictation; cloud with your own keys | 2,000 words a week on desktop, smaller caps on phones |
| Paid plans | Pro $9.99/month or $99.99/year; Local+ $29 once | Pro $15/month or $144/year; Growth from $18 per user a month |
| Cleanup | Modes with your own AI Instructions; optional Local Model Cleanup | Automatic |
| File transcription | Yes, on desktop | No |
| Meetings | Recording on Mac, transcribed afterwards | Notetaker on Mac and Windows |
| Languages | 100+ | 100+ |
| Students | 50% off Pro | 3 months of Pro free, then 50% off |
| Refunds | 14 days on website purchases | Only where the law requires |
Swipe to see all columns
As of October 2026, from spokenly.app, wisprflow.ai and each app's help pages.
Price: Spokenly costs less at every level#
| Spokenly | Wispr Flow | |
|---|---|---|
| Free | Unlimited with local models | 2,000 words a week on Mac and Windows; limited weekly use on phones |
| Pro, monthly | $9.99 | $15 per user |
| Pro, yearly | $99.99 | $144 per user ($12 a month) |
| One-time | Local+ $29 (no cloud models) | None |
| Teams | 30% off Pro per seat, minimum 2: $6.99 a month or $69.99 a year per seat | Pro with no seat minimum; Growth $18 per user a month billed annually ($23 monthly) |
| Students | 50% off Pro: $4.99/month or $49.99/year | 3 months of Pro free, then $6 a month billed annually |
| 3 years, Pro on yearly billing | $299.97 | $432 |
Swipe to see all columns
Prices as of October 2026, from Spokenly pricing, Spokenly teams, Spokenly students, Wispr Flow pricing and Wispr Flow for students.
Spokenly is the cheaper app whichever way you buy it. If local models are enough, it costs nothing; on Pro with yearly billing, it saves about $44 a year. Wispr Flow's advantages are elsewhere: its Pro plan works for small teams with no seat minimum, and its free plan includes cloud transcription and cleanup with no keys or downloads, which Spokenly's free tier doesn't.
A few Wispr Flow details: the free plan's count resets every Sunday at midnight Pacific time, and its help center puts the phone caps at about 750 words a week on iPhone and 1,500 on Android, adding that allowances can vary by account (Wispr Flow help center). The same article refers to a standard 14-day Pro trial.
Refund terms differ sharply. Spokenly refunds website purchases in full within 14 days (refund policy). Wispr Flow's terms say you can cancel at any time, but refunds are only issued where the law requires them (terms of service).
Inside its iPhone app, Spokenly sells Pro through Apple at its own prices: the US App Store lists $9.99 or $10.99 a month and $49.99 or $54.99 for six months. For the full breakdowns, see our Spokenly pricing guide and Wispr Flow pricing guide.
Platforms#
- Mac: both. Spokenly needs macOS 13.3 or later and Wispr Flow macOS 12 or later. Get Spokenly from its website, since its Mac App Store version can't paste automatically and is no longer updated (Spokenly docs).
- Windows: both, on Windows 10 and 11. Wispr Flow needs an x64 processor and doesn't support Windows on ARM (Wispr Flow requirements). Spokenly's Windows app comes from the Microsoft Store or a direct installer, and Spokenly says its Mac and iPhone apps are its most mature (Spokenly for Windows).
- Linux: Spokenly only, with deb, rpm and AppImage builds for X11 and Wayland (Spokenly for Linux). Wispr Flow lists Linux as unsupported.
- iPhone and iPad: both have an iPhone keyboard. Wispr Flow's needs iOS 18.3 or later and doesn't support iPad; Spokenly's App Store listing covers iPhone and iPad from iOS 16.4.
- Android: Wispr Flow only, on Android 13 or later. Spokenly has a waitlist (Spokenly for Android).
- Browser: Spokenly has a Chrome extension (Chrome Web Store).
Features: cleanup, models, meetings and extras#
Cleanup. Wispr Flow cleans up as part of the product, and its data controls page says it uses the app you're dictating in and the text around your cursor to set tone, capitals and punctuation (data controls). Spokenly makes cleanup something you configure. Its modes run AI Instructions you write, through Spokenly's hosted AI on Pro, your own AI provider or Apple Intelligence, and its newer Local Model Cleanup removes fillers and applies spoken corrections on your computer, but it's off by default and strongest in English (Modes, Local Model Cleanup).
Models and keys. Spokenly lets you choose a model for each mode, from local Whisper, Parakeet and Apple's Speech Analyzer to cloud models from OpenAI, Deepgram, Soniox and ElevenLabs, and use your own API keys from 11 providers. Wispr Flow handles transcription in its own cloud.
Editing and actions. Wispr Flow's Command Mode lets you highlight text and say how to change it, on Mac and Windows, with a paid plan or an active trial (Command Mode). Spokenly's Agentic Actions go a different way: on a Mac, a mode can search the web, open or quit apps, or run a Shortcut or shell command after you speak (Agentic Actions).
Meetings. Wispr Flow's Notetaker runs on Mac and Windows without a bot, identifies speakers, answers questions about your meetings and shares notes with AI tools over MCP; the free plan has a weekly meeting limit. Spokenly records your microphone and the Mac's audio, also without a bot, and transcribes the recording afterwards, on Mac only.
Files and developer tools. Spokenly transcribes audio and video files on the desktop, and on the Mac it has a command-line tool and an MCP server that lets Claude Code, Codex and Cursor ask you questions you answer by voice. Wispr Flow can't transcribe files you upload (Wispr Flow help center); its MCP support shares your meeting notes with AI tools instead.
Privacy and security#
Spokenly depends on the path you choose (privacy policy). With local models, audio and transcripts stay on your device, though modes with AI Instructions can send the transcript to an AI service. With Spokenly's cloud models, audio goes through its backend, which doesn't store it, to a transcription provider. With your own keys, audio goes straight to your provider. Local Only Mode blocks network access, analytics included, and can be enforced on company Macs and PCs (Local Only Mode).
Wispr Flow is always cloud-based. Its data controls let you decide whether your audio, transcripts and edits may be used to train its models, and whether your dictations are kept on its servers through Dictation Cloud Storage. It says its third-party AI providers are bound by zero data retention agreements and can't train on your data, and that it may collect usage statistics such as word counts whatever your settings (data controls). For companies it lists SOC 2 Type II and ISO 27001; Growth adds admin control over model training and org-wide HIPAA enforcement with a signed BAA, and Enterprise locks training off (pricing).
The short version: if your audio must never leave the machine, only Spokenly can do that. If you need certifications and admin controls for a company rollout, Wispr Flow lists them on its pricing page.
Where LumeVoice fits#
LumeVoice is our product, so read this as the maker's view. It sits much closer to Wispr Flow than to Spokenly: it's cloud-only and always needs a connection, and it cleans up as you talk, removing filler words and stutters, applying self-corrections and turning spoken lists into real lists. It runs on Mac, Windows, iPhone and Android, in 100 languages, and costs $14.99/mo, $79.99/year or $249 once, after 3,000 free words, no card required. Students get 50% off, and anyone who finds typing hard gets 50% off (accessibility discount).
- Choose it over Wispr Flow if you'd rather buy a license once than subscribe.
- Choose it over Spokenly if you want finished text without setting up models or instructions, or you need Android.
- Don't choose it if your audio must stay on your device or you work on Linux. Spokenly fits both better.
Try LumeVoice on Your Own Words
Hold a key, talk, and LumeVoice types the finished text where your cursor is: in Gmail, Slack, Notion or your code editor. Your audio isn't stored after processing, your history stays on your device, and nothing you say is used to train a model. See pricing or our comparison with Spokenly.
The bottom line#
Spokenly and Wispr Flow suit different people. Spokenly is for people who want control and low cost: free local dictation with no cap, offline use, Linux, your own keys and a cheaper Pro plan. Wispr Flow is for people who want dictation to just work in the cloud, Android included, with meeting notes and the security paperwork companies ask for. Both have free plans, so dictate a few days of real work in each before you pay.
