Aqua Voice
Aqua Voice · Writing · Voice & Language
Aqua Voice is an AI dictation app for Mac, Windows, iPhone, and Android that turns speech into clean text in any app, powered by its own Avalon speech model.
About Aqua Voice
What Is Aqua Voice
Aqua Voice is dictation software built around a speech model the company trained itself, called Avalon. You hold a hotkey, talk, and the words appear in whatever text field has your cursor, with punctuation and casing already handled.
Most voice to text tools live inside one app or one browser tab. Aqua Voice sits at the system level instead, so the same hotkey works in Slack, Gmail, Xcode, a terminal window, or a note taking app without any per app setup.
The selling point is raw speed. The company puts speaking at roughly 230 words per minute against about 40 for typing, and Avalon was trained on developer speech so that commands, model names, and file paths survive the trip into text.
There's a real limit to plan for. Everything runs in the cloud, so there's no offline mode, and the free plan stops at 1,000 words.
Getting Started
- Download the desktop app from the official site and sign in with an email address.
- Approve microphone access plus the system permissions the app asks for on first launch.
- Put your cursor in the target field, hold the hotkey, speak, then release.
- Open settings and add names, brand words, or code terms to the custom dictionary.
- Review transcript history later, or switch on Privacy Mode if you don't want anything stored.
Product Information
A quick look at Aqua Voice's pricing, supported platforms, and performance.
Best for
The users, tasks, and scenarios where this tool fits best.
Users
- Developers and engineers
- Writers and editors
- Students
- Support and sales teams
Tasks
- Prompting AI coding assistants
- Drafting emails and documents
- Capturing meeting notes and voice memos
- Long form writing
Scenarios
- Working inside an IDE or terminal
- Replying on a phone
- Multilingual work
- Hands free or low mobility use
Key features
Avalon Speech Model
Avalon is the model behind every Aqua Voice transcript. On AISpeak-10, the company's own benchmark of clips where people say things like "Claude Code" or "GPT-4o-mini," it scored 97.4% key term accuracy while Whisper Large v3 scored 65.1%. At its October 2025 debut it also ranked first among proprietary models on the public OpenASR leaderboard with 6.24% average word error rate. Those numbers come from the vendor, so treat them as a starting point rather than a neutral verdict.
Dictation In Any App
The desktop app runs at the system level, which means voice typing works in every field that accepts text. There's no plugin per site and no browser extension to keep updated. You hold the hotkey, speak, and release, and the text lands where the cursor sits, including inside a terminal or a code editor.
Custom Dictionary
You can teach the app the words it keeps getting wrong: colleague names, product names, internal jargon, or library names. Accuracy on those terms improves right away, which matters most for people whose work vocabulary is unusual. It's the fastest fix for the one category of error that speech to text tools usually make.
Writing Style Rules
Instead of one fixed output format, you set the tone and rules once and the app applies them to every transcript. Lowercase for Slack, full sentences for client email, structured headings for documentation. The text arrives closer to publishable, so editing time drops.
49 Language Support
Aqua Voice understands 49 languages and adapts to how you actually speak, including accent and pacing. You can dictate in one language and keep your notes in another, and mixed language sentences are handled better than in tools built around a single locale.
Transcript History And Privacy Mode
Every session is saved to history by default, so you can search back for a paragraph you spoke last week. Privacy Mode flips that: nothing is stored on the company's servers at all. If your work involves confidential material, that switch is what makes the tool usable.
Avalon API
Developers can call the same model through the Avalon API, and the docs describe a two line swap of base URL and model name to move an existing pipeline over. That makes it possible to keep your request shape and auth while getting better recognition on technical speech.
Pros and cons
Pros
- Key term accuracy on technical speech is the strongest part of the product, and the AISpeak results back that up for developer use.
- One hotkey covers every app on Mac, Windows, iPhone, and Android, so there's no per app integration to maintain.
- The free tier needs no credit card and gives a real trial at 1,000 words rather than a locked demo.
- Security paperwork exists for teams: SOC 2 Type II, ISO 27001, plus GDPR and HIPAA compliance claims.
Cons
- There's no offline mode. On a plane, in a locked down network, or during an outage, dictation stops working entirely.
- The free plan ends at 1,000 words, which is roughly a few pages, so most regular users hit the paywall within a week.
- The $10 and $19 monthly rates require annual billing. Month to month costs $12 and $24, and there's no lifetime option.
- Accuracy claims come from benchmarks the company built itself, so independent testing is still thin.
Frequently asked questions
Yes, there's a free plan with 1,000 words and access to the Avalon 1.5 model, and it doesn't ask for a card. Past that limit you need Pro at $10 per month billed annually, or $12 month to month.
Related content
Explore related tools, skills, and articles for Aqua Voice.
Aqua Voice Alternatives
Google EmbeddingGemma 2
Google DeepMind · Writing · Image · CodingGoogle EmbeddingGemma 2 is a 740M-parameter open model that turns text, code, images, audio, and video frames into vectors in one 768-dimension space, running on phones and laptops.
MePersona
MePersona, Inc. · Writing · Productivity · MarketingMePersona is an AI unified inbox for creators. It learns your voice, drafts replies to fan messages from 12 platforms, and sends only what you approve.
Fluence
Yay Software (Yannick Huchard) · Writing · Voice & Language · ProductivityFluence is an Android app that reads PDFs, EPUBs, Word files and web articles aloud, follows each sentence as it speaks, and answers questions about the text on the phone itself.