See the difference, live
Aqua Voice refines your words as you speak: punctuation, formatting, and technical terms landed in real time.
Aqua Voice vs Handy summary
Aqua Voice and Handy both dictate into any app with a hotkey, on opposite philosophies. Handy is free and open-source, running open models (Whisper, Parakeet) locally on Mac, Windows, or Linux. Aqua Voice runs its own model, Avalon (97.4% on technical terms), in the cloud, with zero setup plus iPhone. Pick Handy for free and offline; Aqua Voice for accuracy and convenience, $10/mo billed annually.
You want managed dictation with published model results with nothing to set up; you dictate code, AI prompts, or technical terms and want accuracy you can verify; you want real-time cloud processing, iPhone and Android support, and ongoing improvements; or you'd rather not manage a local model yourself.
You want a free, open-source tool; you need everything to run on-device and offline; you're on Linux; or you like reading the code, swapping models, and self-managing your setup.
What models does Handy use?
Handy is open about its models: it runs OpenAI's Whisper (via whisper.cpp) and NVIDIA's Parakeet locally, and you pick which one. These are strong, general-purpose models, freely available and well-documented.
The tradeoff with general-purpose models is that they weren't designed for specialized vocabulary. Everyday speech is fine, but technical terms, programming keywords, and AI tooling are where they struggle. That's the gap a purpose-built model closes, so we trained our own and published its results.
We evaluated Avalon on a dataset of real clips of developers and AI researchers talking naturally, and measured it against Whisper Large v3, one of the open models Handy can run:
| Model | AISpeak-10 key-term accuracy |
|---|---|
| Avalon (ours) | 97.4% |
| Whisper Large v3 | 65.1% |
On the public OpenASR leaderboard, Avalon ranked #1 among proprietary systems at its October 2025 debut, ahead of Whisper on that leaderboard. Handy also offers NVIDIA's Parakeet, which we haven't benchmarked head-to-head; either way, the technical-vocabulary gap is what matters most to developers and power users. Try Aqua Voice free.
AISpeak-10 is Aqua’s English technical-speech dataset. Avalon transcribed the key term correctly in 97.4% of clips, versus 65.1% for Whisper Large v3. This measures key-term recognition, not overall transcript accuracy or a head-to-head test of these apps. See the evaluation and results.
Aqua Voice vs Handy: Feature-by-feature comparison
| Feature | Aqua Voice | Handy |
|---|---|---|
| Speech model | Avalon (own, purpose-trained) | Open models (Whisper, Parakeet), local |
| Cost | $10/mo ($120/yr), free tier | Free, open-source (MIT) |
| Public benchmark | #1 proprietary on OpenASR (Oct 2025 debut) | Not published (model-dependent) |
| Technical-term accuracy | 97.4% (AISpeak-10, our benchmark) | Not published (model + hardware) |
| Setup | Install app; no local model download | Download and pick a local model |
| Processing & security | Cloud; SOC 2 Type II audit report, ISO 27001, GDPR, HIPAA | On-device |
| Works offline | ❌ (cloud) | ✅ (on-device) |
| Mac / Windows | ✅ | ✅ |
| Linux | ❌ | ✅ |
| iPhone / Android | ✅ | ❌ |
| Real-time + screen context | ✅ | Hardware-dependent |
| Account required | ✅ | ❌ (no account) |
| Best for | Accuracy, zero setup, technical/AI users, iPhone, Android | Free, on-device, open-source, Linux |
Speech model
Cost
Public benchmark
Technical-term accuracy
Setup
Processing & security
Works offline
Mac / Windows
Linux
iPhone / Android
Real-time + screen context
Account required
Best for
Where Aqua Voice is better
Accuracy without the setup
Avalon is purpose-trained for technical and AI speech and benchmarked publicly: 97.4% on AISpeak-10 where Whisper Large v3 scores 65.1%, and #1 among proprietary systems on the OpenASR leaderboard at its October 2025 debut. Handy's accuracy depends on which open model you download and how powerful your machine is.
Zero setup, always current
Aqua Voice runs server-side, so there are no models to download, no hardware requirements, and no configuration. The model keeps improving on its own, where a local tool is frozen to whatever you've installed until you update it.
Real-time on a bigger model
Because Avalon runs in the cloud, it streams text as you speak using a model hosted on managed servers. On-device tools are gated by your machine: the more accurate the local model, the slower it goes.
iPhone and Android and screen context
Aqua Voice runs on iPhone and Android as well as Mac and Windows, on one subscription, and the desktop app reads the active window for context, code-aware in your editor, casual in Messages. Handy is desktop-only (Mac, Windows, Linux).
A model developers can build on
The same model is available through the Avalon API, OpenAI-compatible, at $0.39/hour of audio, so you can put Avalon-grade accuracy into your own tools.
Where Handy wins
Free and open-source. Handy is MIT-licensed and completely free, with the source on GitHub. If you want to read the code, self-host, or avoid a subscription entirely, that's a real advantage Aqua Voice doesn't offer.
Fully on-device and offline. Handy runs the model locally, so your audio never leaves your machine and it works with no internet connection. If on-device privacy is non-negotiable, Handy has the edge.
Linux and full control. Handy runs on Linux (Aqua Voice doesn't), and because it's open-source you can swap models and customize it. The flip side is that you manage the model and setup yourself.
Free and on-device vs managed and cloud
Handy is a free, open-source (MIT) dictation app from solo developer CJ Pais. You download it, it pulls open speech models onto your machine, and everything runs locally on Mac, Windows, or Linux. Your audio never leaves the device, there's no account, and there's nothing to pay.
Aqua Voice takes the opposite approach. We run our own model, Avalon, in the cloud, so there's no local speech model to download or configure, the model is hosted on managed servers, and it keeps improving without you updating anything. The tradeoff is honest: it's a paid, cloud service rather than a free, on-device one.
How to decide
Aqua Voice is free to start, so you can compare it against Handy on the words you actually dictate before paying anything.
Or just ask ChatGPT.
Frequently asked questions
Is Aqua Voice a good Handy alternative?
It depends on what you value. Handy is free and open-source, and runs entirely on your own machine. Aqua Voice is a managed service on its own model, Avalon, with published technical-term results (97.4% on AISpeak-10), zero setup, iPhone and Android support, and ongoing improvements. If you want a free, on-device, do-it-yourself tool, Handy is great; if you want managed dictation with published model results with nothing to configure, Aqua Voice is the stronger choice.
What's the difference between Aqua Voice and Handy?
They're built on opposite philosophies. Handy is free and open-source (MIT), running open speech models like OpenAI's Whisper and NVIDIA's Parakeet locally on your Mac, Windows, or Linux machine. Aqua Voice is a managed cloud app running Avalon, its own purpose-trained model, with real-time streaming and screen-context awareness on desktop, plus iPhone and Android support. Both dictate into any app with a hotkey.
Is Aqua Voice more accurate than Handy?
We have not established an app-to-app accuracy advantage. Handy runs general-purpose open models (Whisper, Parakeet), and its accuracy depends on which model and hardware you choose. Aqua Voice runs Avalon, trained specifically for technical vocabulary: it scores 97.4% on AISpeak-10 (our benchmark of AI and technical terms) where Whisper Large v3 scores 65.1%, and it ranked #1 among proprietary systems on the public OpenASR leaderboard at its October 2025 debut.
Is Aqua Voice free like Handy?
Handy is fully free and open-source. Aqua Voice has a free Starter tier (1,000 words), then Pro is $10/month billed annually. You're paying for a purpose-built, benchmarked model, zero setup, real-time cloud processing on desktop, iPhone and Android support, and ongoing model improvements and support, rather than running and maintaining an open model yourself.
Does Aqua Voice work offline like Handy?
No. Handy runs on-device, so it works offline and your audio never leaves your machine. Aqua Voice processes audio in the cloud to run the Avalon model, so it needs an internet connection. If fully offline, on-device dictation is a requirement, Handy is the better fit.
Does Aqua Voice run on Linux?
No. Aqua Voice runs on Mac, Windows, iPhone, and Android. Handy also supports Linux, so if you need Linux dictation, Handy is the option. Aqua Voice covers iPhone and Android, which Handy does not.
Why pay for Aqua Voice when Handy is free?
For accuracy and convenience. Avalon is purpose-trained and benchmarked publicly: on AISpeak-10 it scores 97.4% to Whisper Large v3's 65.1% (Whisper is one of the open models Handy runs). Aqua Voice also requires zero setup (no model downloads or tuning), runs Avalon on managed servers, adds screen-context awareness and iPhone and Android support, and keeps improving with ongoing updates and support. Handy is an excellent free, on-device option if you'd rather self-manage an open model.
Is Aqua Voice good for coding and AI prompting?
Yes, it's what Aqua Voice is built for. Avalon was trained on the speech developers and AI users produce, reaching 97.4% accuracy on technical terms like programming keywords, framework names, and CLI commands. It also reads the active app for context and offers the Avalon API for developers.
