Features
The right AI model for every job
Transcript.audio uses leading AI models from Anthropic, Mistral, Qwen and OpenAI. Use the recommended set, or choose exactly which model does each job.
All plans
How it works
Choose a preset: Recommended, Most accurate, Cheapest, Fastest, Privacy optimised or Free. Each one picks the best model it can for every job.
Or choose by hand for each job: transcription, meeting notes, details to capture, titles and folder names, and the help assistant. You see what each model is good at before you switch.
Every job has backup models lined up, so if one is busy the next one takes over and your notes still arrive.
What you get
Choose the AI for each job
Pick a preset, or choose the model for transcription, notes, details, titles and help separately.
Put the most accurate model where it matters and the fastest everywhere else.
Top AI models included
Notes written by leading models such as Claude, with AI included in your plan.
Better models mean fewer corrections, and there are no API keys or AI bills to manage.
Premium and Pro Converser
Learn more: Top AI models includedPrivacy optimised mode
Transcription and notes run entirely on your computer, so nothing leaves it.
The strongest privacy there is for confidential conversations, on every plan.
All plans
Learn more: Privacy optimised mode
Frequently asked questions
Do I need my own API keys?
No. AI is included in your plan, so there is nothing to set up.
Which models are available?
Models from Anthropic (Claude), Mistral, Qwen and OpenAI, plus on-device models in privacy optimised mode. The models page lists them all.
Can I go back to the recommended choice?
Yes. One click puts any job back on the preset.
Related features
Try it on your next meeting
Download free and your first meeting notes are minutes away. No bot, nothing to set up.
Try Pro Converser free for 1 month. All future updates included, free.