Viaalink Notetaker vs Real-Time Voice Cloning
Real-Time Voice Cloning is not a meeting notetaker but a research project (a master's thesis): it turns a few seconds of audio into a digital voice that speaks any text. Notetaker does the reverse: it turns spoken words into text.
Where Notetaker is stronger than Real-Time Voice Cloning
Phone and computer
Android, and Windows (alpha), macOS and Linux. The phone sends a session to your computer over your own Wi-Fi.[2]
No bot in your call
You start Notetaker yourself on your phone or computer. Nobody extra joins.[1]
Everything stays on your device
Speech becomes text on your own phone or computer. Audio, transcripts and reports never reach Viaalink.[2]
99 languages
On your phone Notetaker recognises speech in all 99 languages Whisper knows. We measured the accuracy for Dutch and English.[3]
Transcribe existing recordings
Also audio and video files you already have, on your phone and on your computer.[4]
Voice profiles with consent
Can also be made afterwards from a recording, and they stay on the device.[5]
Side by side
Only what works today. Where their site is silent, we say so.
| Topic | Viaalink Notetaker | Real-Time Voice Cloning |
|---|---|---|
| A bot in your call | No bot You start Notetaker yourself on your phone or computer.[1] | Not stated on their site[9] |
| Where speech becomes text | On your device On your own phone or computer.[2] | Not stated on their site[9] |
| Where recordings and text are kept | On your device Audio, transcripts and reports never reach Viaalink.[2] | Not stated on their site[9] |
| What writes the report | Your phone A language model on the phone itself. The report never reaches Viaalink.[2] | Not stated on their site[9] |
| Phone and computer | Both, together Android, and Windows (alpha), macOS and Linux. The phone sends a session to your computer over your own Wi-Fi.[2] Notetaker is ahead here | Computer only Windows and Linux, as a Python project.[9] |
| Without internet | Yes After activation; internet only to fetch the speech model once, to activate and renew your license, and, if you agree, to check for updates.[6] | Not stated on their site[9] |
| Languages | 99 languages Phone and computer: all 99 Whisper languages. Measured for Dutch and English.[3] | Not stated on their site[9] |
| Voices and speakers | On the device Only with consent; on your phone also afterwards from a recording. Recognition with a profile is in beta, separating voices without one in alpha.[5] | Voice cloning A voice from a few seconds of audio.[9] |
| Existing recordings | Yes Audio and video files, on phone and computer.[4] | Not stated on their site[9] |
| Export | Yes Transcript and report, as PDF, Markdown or Word.[7] | Not stated on their site[9] |
| Account needed | For the license The activation code is on your account page.[8] | Not stated on their site[9] |
Coming soon: reports on the computer with AI models on your own machine, your own key for an AI service, and connecting Claude, ChatGPT or Mistral to your phone and computer.
Sources
- viaalink.com/notetaker#how-it-works
- viaalink.com/legal/privacy#viaalink-notetaker
- viaalink.com/notetaker#languages
- viaalink.com/notetaker#import
- viaalink.com/notetaker#voice-profiles
- viaalink.com/notetaker#privacy
- viaalink.com/notetaker#export
- viaalink.com/notetaker#get-notu
- github.com/corentinj/real-time-voice-cloning, read on 12 November 2026
Real-Time Voice Cloning is a trademark of its owner. We only compare what their own pages say, on the date shown.
Try Notetaker yourself
Notetaker is in beta. See for yourself how it works in your own conversations.
Sign up as a tester Buy and sponsor Ask a question
During the beta period we are still analysing our pricing.