Parlo vs Spokenly: which dictation app fits
Spokenly lets you choose a local model, a cloud model, or your own API key, and it ships an iPhone keyboard. Parlo processes speech on-device only, with no model choice to make. Pick Spokenly if you want that flexibility. Pick Parlo if you want one on-device path and a free beta with no account.
Spokenly gives you a choice each time you dictate: a local model that runs entirely on your device, a cloud model, or your own API key sent straight to a provider you pick. Parlo does not offer that choice. Every transcription and cleanup step runs on-device with Apple's speech engine. There is one path, not several.
Spokenly's privacy page describes its cloud mode separately from its local mode. The behavior you get depends on which one is active when you dictate. Parlo's behavior is the same on every dictation, on iPhone and on Mac. It uses Apple's SpeechAnalyzer and SpeechTranscriber for transcription and an on-device language model for cleanup.
| Parlo | Spokenly | |
|---|---|---|
| Platforms | iPhone, Mac | Mac, Windows, Linux, iPhone |
| Where processing happens | On-device only | Local models on-device, or cloud models sent to its backend, by your choice |
| Account required | No | No for the free local tier; yes for Pro |
| Price | Free during the beta | Local models free; Pro about $8.33 a month shown rate, or $99.99 a year |
| Free tier | One version, free during the beta | Local models, unlimited use, per its pricing page |
| Dictation history | Local, encrypted; end-to-end encrypted iCloud sync on by default | not stated |
| Export | Markdown, Mac only | not stated |
| iPhone keyboard | Yes, its own keyboard extension | Yes, a custom dictation keyboard |
| Mac hotkey | Yes, hold Option | not stated |
| Custom dictionary | Yes, syncs to the keyboard | Yes, plus a separate word-replacement feature |
| Languages | English (beta) | 100 plus, with automatic detection |
Spokenly's site also describes a "Local Only Mode" that can restrict network access for full offline use on top of its local models. That is a second, stricter setting layered on the first choice. Parlo has one behavior, always on-device, with no separate offline mode to turn on.
What Spokenly does better
Spokenly gives you a model choice each time you dictate, including a fully local option, a cloud option, and a bring-your-own-key option that sends audio straight from your device to the provider you pick. It also runs on Windows and Linux, which Parlo does not. It adds a word-replacement feature next to its custom dictionary.
What Parlo does better
Parlo has one processing path: on-device, every time, with no model or provider to choose. There is nothing to configure and no cloud mode to accidentally leave on. Parlo's history is encrypted at rest on the device. If sync is on, it moves through your private iCloud end-to-end encrypted. Parlo is also free during the beta with no account and no subscription tier to weigh against a free one.
Who should pick which
If you want the option to switch between a local model and a cloud model, Spokenly's flexibility covers cases Parlo does not touch. The same is true if you already run Windows or Linux. If you would rather have one on-device path with nothing to pick between, plus a free beta and no account, Parlo keeps that simpler.
Clinicians and lawyers who want to avoid any cloud transcription mode entirely may prefer Parlo's single on-device path over toggling a setting correctly every time. There is no mode to leave on cloud by accident. Founders and operators who dictate on both iPhone and Mac and want one consistent behavior across both may also prefer Parlo. Parlo's custom dictionary lives on iPhone and feeds the keyboard's autocorrect. Spokenly's site does not say whether its dictionary syncs across its apps. Writers who want a Markdown export on Mac get that from Parlo directly. Spokenly's site does not name an export feature at all.
Spokenly's Pro plan also carries a 14 day refund window and can be canceled anytime, per its pricing page. Parlo has no purchase to refund since the whole beta is free. Its bring-your-own-key option means Spokenly itself charges nothing for that path, though the provider you choose bills you directly. That is a cost Parlo's on-device path never introduces.
People doing hands-free dictation for accessibility reasons may notice that Spokenly's site does not name a Mac hotkey. Parlo for Mac uses a hold-Option gesture with a double-tap lock for continuous hands-free use. On iPhone, the build 54 accessibility pass also improved tap targets and VoiceOver labels on Parlo's keyboard.
Switching notes
If you use Spokenly's local-model mode today, the processing model is close to Parlo's: on your device, no cloud step. The switch looks like this:
- Install Parlo for iPhone from TestFlight, or download Parlo for Mac from the beta page, or both, all free during the beta.
- Turn on the Parlo keyboard in iPhone Settings and allow Full Access.
- Re-add the names, terms, and word replacements you set up in Spokenly to Parlo's custom dictionary. None of that carries over.
If you rely on Spokenly's cloud mode or your own API key for a specific provider, Parlo does not offer an equivalent. It transcribes and cleans up on-device only. Reply to beta@parlokey.com with anything that looks wrong along the way, since Parlo is a beta and that feedback shapes the next build.
Frequently asked questions
Does Spokenly work without an account?
Its pricing page states no account or subscription is required for the free local-model tier. Its Pro tier needs sign in and payment.
Does Parlo need an account?
No. There is no Parlo account and no server of ours holding your content.
Which app has an iPhone keyboard?
Both. Spokenly's home page names a custom dictation keyboard for iOS, and Parlo ships its own iPhone keyboard extension.
Can Spokenly send audio to a server?
Its own privacy page describes a cloud mode where audio is sent to its backend for transcription, separate from its local-model mode.
Does Parlo offer a cloud option?
No. Parlo only transcribes and cleans up text on-device, with Apple's speech engine and no cloud model choice.