Which transcription entry points are supported?

Audio Note supports file transcription, realtime microphone capture, realtime app capture, global realtime subtitles, link transcription, and watch-folder automation.

Is my data uploaded to the cloud?

The default transcription workflow runs locally, so your audio and transcripts stay on your device. Content is only sent out when you intentionally use a cloud AI feature.

Which languages are supported?

Audio Note supports Chinese, English, Japanese, Korean, Russian, and many other languages, with automatic detection when the selected model supports it.

How should I choose models on a lower-performance device?

The app includes device benchmarking and model recommendations. Older machines can start with lighter models or Sherpa for realtime capture, while faster machines can switch to higher-accuracy options.

Which NVIDIA GPUs work with CUDA on Windows?

GTX 10-series and newer NVIDIA cards can use the CUDA engine for Whisper GPU acceleration. Older cards may crash on Vulkan—the app will suggest switching to CUDA in Settings → Transcription → GPU engine. See the docs page “NVIDIA GPU & CUDA compatibility” for details.

What are advanced ASR models (Pro)?

They are Pro-only offline file models under Settings → Models → ML Engine (e.g. Qwen3-ASR, Cohere, Parakeet). Use them when you want that offline quality route on Pro—not for low-latency live capture (see realtime models).

What is included in a backup file?

Backups include notes, transcription text, task history, AI chats, and common settings. They do not include raw recordings, downloaded models, or exported caption files. Create encrypted backups under Settings → Backup & Restore—see the “Backup & Restore” doc for details.

Do I need to use the Internet when transcribing?

For local audio/video files, link transcription, and realtime transcription, the transcription engine runs on your device. Account, authorization, subscription, payment, and feedback features connect to Audio Note servers when you use them.

Audio Note - Local Realtime Transcription for Mic, Apps, System Audio, and Files

Every word lands in a notebook.

Meetings, podcasts, and pasted links become one editable note — not a raw dump of text.

00:02

Launch date moved to next Tuesday. Alice owns the checklist.

Four ways in, one notebook

Come in from a file, live audio, a link, or a watched folder — every transcript lands in the workspace.

Local file

File transcription

Drop local audio or video, keep your preset visible, and let the local engine write an editable note.

Audio and video in one queue — the player stays in sync with the note
Presets for meetings, interviews, and captions keep settings one click away

PresetsTimestampsAudio & videoGPU optional

Q2 Launch Board Review.mp4Investor Podcast Cut.m4a · QueuedWeekly Product Sync.m4a · DoneBatch queue · 3 files leftTranscribe · Small · EnglishQ2 Launch Board Review.mp4Investor Podcast Cut.m4a · QueuedWeekly Product Sync.m4a · DoneBatch queue · 3 files leftTranscribe · Small · English

Re-run

Q2 Launch Board Review.mp4

00:00 / 09:28

Live

Realtime transcription

Capture microphone, application audio, or system audio and turn the stream into live captions.

Lines appear as multiple previews, then commit — and Live Refine fixes wrong words in place
Live translation runs under every line by default

Live refineLive translationPresetsMic / app / system

In Progress · 02:14APAC needs the captionsTranslation · 中文Live refine · Shipping TuesdayMicrophone · Meeting presetIn Progress · 02:14APAC needs the captionsTranslation · 中文Live refine · Shipping TuesdayMicrophone · Meeting preset

Realtime transcription

ModelSmall · Realtime

LanguageEnglish

MicrophoneMacBook Pro Microphone

SceneMeeting

Listening for speech…

Preview text shows up before each line is committed.

TranslationLive Refine

In Progress· 02:18Stop

Queue

Link transcription

Paste a supported link, download media locally, and generate a transcript with task history.

YouTube and Bilibili links download and transcribe without leaving the app
Every job keeps its own status and history, so retries never lose track

YouTubeBilibiliLocal downloadTask history

youtube.com/watch?v=q2-launch0%bilibili.com/video/BV1q2 · Download queued · 2podcast/episode-12 · Completed2 concurrent downloadsArchive after transcriptionyoutube.com/watch?v=q2-launch0%bilibili.com/video/BV1q2 · Download queued · 2podcast/episode-12 · Completed2 concurrent downloadsArchive after transcription

Paste a linkStart

Paste a YouTube or Bilibili link to transcribe locally.

Automatic

Watch folder

Monitor a folder continuously. New files are detected, queued, transcribed, and archived.

Video and audio files are converted and transcribed automatically, hands-free
Live task stats show what is converting, transcribing, and done

Auto queueFolder monitorArchiveUnattended

Recordings/standup-0912.m4a · Transcribingall-hands-0909.m4a · Donecustomer-call-0905.m4a · DoneNew file detectedAuto export · SRTRecordings/standup-0912.m4a · Transcribingall-hands-0909.m4a · Donecustomer-call-0905.m4a · DoneNew file detectedAuto export · SRT

Folder WatchAdd Watch

File Browser9 files

Search files…

Recordings2

keynote-0828.m4a

webinar-0907.mp3

all-hands-0909.m4a

customer-call-0905.m4a

interview-raw-0903.wav

demo-capture-0902.mp4

podcast-cut-0901.mp3

Client Calls1

acme-kickoff-0904.m4a

beta-interview-0830.wav

Podcast Archive

3

Processing Tasks

13

Completed

1

Failed

Active TasksHistory Tasks

keynote-0828.m4aTranscribing64%

beta-interview-0830.wavConverting8%

webinar-0907.mp3Pending0%

Pick a job, get a notebook

One tool that fits every recording job you actually do.

Record from the built-in mic or any input — one take, saved locally and ready to transcribe. Read the docs

Recording

Ready to record

00:00

MacBook Pro Microphone

Record

Watch a full file transcription

From opening a file to editing the finished transcript, in under two minutes.

Pricing

No surprise cloud fees. Local compute is on you; AI help is metered.

Compare all features Models & engines guide (docs)

Standard

Suitable for people who need to transcribe larger audio and video files into text and use AI functions

$9

$7/month

Save more with yearly billing — see the pricing page

Audio, video, microphone recording, and app recording transcription

Microphone, app, and global realtime transcription; dictation coming soon

Translation and real-time translation

Whisper official models (11)

Realtime models

Export captions in 10 formats (Text/SRT/VTT/...)

GPU acceleration and advanced transcription settings

Realtime preset tuning enabled (custom preset parameters unavailable)

Realtime Refine (Beta) — AI correction on finalized transcription lines, uses AI Chat tokens

Watch folders up to 5 with auto export

Link transcription multi-link queue with 2 concurrent downloads

Link transcription file size up to 10GB with unlimited download speed

1 million AI Chat tokens per month

Pro

For people who have a discrete graphics card and want higher transcription speed and accuracy

$12

$9/month

Save more with yearly billing — see the pricing page

Includes all Standard features

Speaker diarization — automatically tells speakers apart

Vocal separation — extract vocals from audio and video

Voice enhancement with DTLN denoising for noisy environments

Watch folders up to 20, including recursive and hidden folders

Link transcription parallel mode with 5 concurrent downloads

All Whisper models (official + community)

All realtime models (Chinese/English/French/Russian/Korean/Japanese)

Advanced ASR models (Qwen3-ASR, Cohere multilingual, Parakeet, … under Settings → Models → ML Engine)

Custom realtime parameter configuration

Realtime app video archive (Pro)

2 million AI Chat tokens per month

Lifetime

Suitable for people who need long-term use and free enjoyment of all subsequent features, one-time purchase, lifetime use

$139/lifetime

All Pro transcription capabilities included

One-time payment, no recurring billing

2 million AI Chat tokens per month for 24 months

Future Pro-level local transcription features included

Free

Small audio files, six lightweight Whisper models — download and start right away.

Download Now

Audio transcription up to 1GB per file

Tiny / Tiny English / Base / Base English / Small / Small English Whisper models

Text format export (.txt)

CPU transcription only

Link transcription single-link mode with 1GB file cap, 1 MB/s limit, and 1 retry

Supports 1 watch folder

No video / recording / realtime / translation features

New users get a 7-day free trial in the app, no credit card required.

Download Audio Note

Mac & Windows. Install it and transcribe locally — no sign-up required.

See all downloads Other builds & system requirements

Guides

From installation to file, realtime, link and folder monitoring.

Read the docs

Frequently Asked Questions

Ready to transcribe?

Download Audio Note free — no account needed for local transcription.

Download free View pricing

Capture speech from mic, apps & media files in real time

Every word lands in a notebook.

Four ways in, one notebook

File transcription

Realtime transcription

Link transcription

Watch folder

Pick a job, get a notebook

Watch a full file transcription

Pricing

Standard

Pro

Lifetime

Free

Download Audio Note

Guides

Frequently Asked Questions

Which transcription entry points are supported?

Is my data uploaded to the cloud?

Which languages are supported?

How should I choose models on a lower-performance device?

Which NVIDIA GPUs work with CUDA on Windows?

What are advanced ASR models (Pro)?

What is included in a backup file?

Do I need to use the Internet when transcribing?

When should I use realtime transcription versus file transcription?

Which export formats are available?

Can I keep editing and organizing the transcript afterward?

What are link transcription and watch folders good for?

Is there a free trial?

How does Audio Note protect data security and privacy?

Does AI Chat send data to the cloud?

How can I get technical support or report issues?

Ready to transcribe?