Mainly for Wispr Flow users who dictate in noisy places; its accuracy claims come only from Wispr's own tests.
Kenning's take
What it does
Canto is a speech model from Wispr's Advanced Interfaces Lab for real-time dictation in Wispr Flow. Wispr says it had the lowest word error rate on its own test set of real dictations, including noisy and whispered speech, when compared with models from Google, OpenAI, AssemblyAI and Deepgram. Those results are Wispr's own.
- Real-time dictation model
- Trained for background noise, whispers and far-field speech
- Tested on 10 hours of real English dictations
- Tied for lowest error rate on LibriSpeech in Wispr's tests
Who it suits. Wispr Flow users who dictate in busy or noisy places.
Key facts
- Category
- AI models
- Part of
- Wispr Flow
- Made by
- Wispr Flow (Wispr Advanced Interfaces Lab)
- Released
- 17 September 2026
- How to get it
- Used for dictation in Wispr Flow; the announcement does not say which plans or platforms use it
- Works on
- Wispr Flow
- Free plan
- Wispr Flow has a free download; Canto plan availability not stated
FAQ
When was Wispr Canto released?
Wispr announced Canto on 17 September 2026.
How do I use Canto?
It is the speech model behind dictation in Wispr Flow. The announcement does not say which plans use it.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.