Short answer: No, Aqua Voice does not transcribe uploaded files - it is a live dictation tool. You hold a hotkey, speak, and the text lands wherever your cursor is. There is no upload button, no import screen, no “drop your recording here” anywhere in the product. If you have a podcast episode, a meeting recording, a voice memo, or an MP4 and you want a transcript from it, Aqua Voice is not the tool for that job.
This trips people up because “voice to text” describes two completely different products. One types for you while you talk. The other turns a file you already have into a document. Aqua Voice is firmly the first kind.
What Aqua Voice actually is
Aqua Voice is a live dictation app for Mac, Windows, and iOS. The workflow is simple: you set a hotkey, hold it, and speak. Aqua transcribes in real time and inserts the text into whatever text field you are working in - Slack, Gmail, your IDE, a browser box, anything with a cursor.
The whole point is replacing keyboard typing in the moment. It is genuinely good at this. Aqua runs its own cloud speech model, Avalon, launched in August 2025 and tuned heavily for technical and coding vocabulary. Aqua claims 97.4% accuracy on its own AISpeak-10 benchmark versus 65.1% for Whisper Large-v3. That is a self-reported figure on Aqua’s own benchmark, so treat it as a vendor claim rather than an independent result - but the direction is real: developers who dictate code and technical terms tend to rate Avalon highly.
That is the important framing here. Aqua is a strong live dictation tool. It is just a different category from file transcription.
Can Aqua Voice transcribe files? (No, and here is why)
There is no file input in Aqua Voice at all. To transcribe a recording, a tool needs a way to accept a file - an upload dialog, a drag-and-drop zone, a “transcribe existing audio” menu. Aqua has none of these. Its interface is built entirely around the live-capture-at-your-cursor loop.
This is confirmed on Aqua’s own FAQ and echoed in competitor comparisons - superwhisper’s own writeup describes Aqua Voice as live dictation only. It is not a hidden setting or a paywalled feature you unlock on the Pro plan. The product simply does not do it.
So concretely, none of these work in Aqua Voice:
- Dropping in a recorded interview to get a transcript
- Uploading a saved Zoom or Teams meeting recording
- Turning a podcast episode into show notes
- Extracting captions from an MP4 or MOV video
- Transcribing an old voice memo off your phone
If your task is any of those, you need a file-transcription tool, not a dictation tool.
The other catch: Aqua is cloud-only
Even setting files aside, there is a second thing worth knowing. Aqua’s FAQ states plainly that it is cloud-based and needs a connection. Every dictation you make is sent to Aqua’s servers, where the Avalon model runs. There is no on-device or offline mode at any tier.
For a lot of people that is fine. For anyone dictating client information, legal notes, medical content, or anything under a privacy policy, it means your spoken audio is leaving your machine every single time. That is a reasonable dealbreaker in some workflows, and it is worth naming clearly rather than burying it.
For reference, Aqua’s pricing is $8/month billed annually ($96/year), with the iOS App Store tier running pricier at $12.99/month. The free tier is a one-time 1,000 words for the lifetime of the account, roughly 8 minutes of speech.
Aqua Voice vs SnailText
If you want live dictation, Aqua is excellent and you should try it. But if you also need to transcribe recordings you already have, locally, that is a job Aqua does not do. Here is the honest side-by-side.
| Aqua Voice | SnailText | |
|---|---|---|
| Live dictation | Yes - its core feature | Yes |
| Transcribe uploaded files | No | Yes - drag-drop files and folders |
| Where audio is processed | Aqua’s cloud (always) | On your device |
| Works offline | No | Yes |
| Speech model | Avalon (proprietary, cloud) | Whisper + Parakeet (local) |
| Languages | Multiple | 99 |
| Subtitle export (SRT/VTT) | No | Yes - plus TXT, MD, JSON |
| Speaker diarization | No | No |
| Platforms | Mac, Windows, iOS | Mac, Windows |
| Free tier | 1,000 words, one-time | Unlimited with compact models |
| Pro pricing | $8/mo annual ($96/yr) | $7.49/mo · $89/yr |
The line that matters: a dictation-only tool has no way to produce an SRT or VTT subtitle file, because it never handles a finished recording with timestamps. SnailText handles both the live-typing loop and the drop-in-a-file loop, and it runs the speech model on your own hardware, so the audio never leaves your machine.
To be fair to Aqua: its Avalon model has a self-claimed accuracy edge on technical and coding vocabulary that a local model on a consumer laptop will not always match. If pure live-dictation accuracy on dev terms is your only concern, that edge is worth weighing.
Honest verdict
Aqua Voice does not transcribe audio files - not at Pro, not at any tier, and no 2025 or 2026 update changed that. It is a live dictation tool, and a good one, but it only converts speech you say in real time, and it does it in the cloud.
So the decision is clean:
- You want fast, accurate live dictation and cloud is fine? Aqua Voice is excellent. If you are weighing it against other dictation tools, see our Aqua Voice alternatives roundup.
- You need to transcribe existing recordings, locally, without uploading? That is a different job, and Aqua cannot do it. SnailText transcribes files on your own machine - Whisper and Parakeet running on-device, 99 languages, SRT and VTT export, and a free tier that covers file transcription with the compact models.
If you want both live dictation and local file transcription in one app, try SnailText - Mac and Windows, nothing uploaded, free to start.
Aqua Voice product details and pricing from aquavoice.com and its public FAQ, mid-2026. The 97.4% accuracy figure is Aqua’s own self-reported benchmark result, not an independent test. Products update pricing and features periodically - check aquavoice.com for current details.
Related reading:
- Aqua Voice alternatives - the full roundup if you are shopping for a dictation tool
- Transcribe audio files locally - drop in recordings and get transcripts without uploading anything