Capture Speech from Mic, Apps & Media Files in Real Time
Whether it's transcribe audio/video files, microphone audio, or real-time transcribe apps, Audio Note will transcribe to text exactly for you locally.
10K+ people are using it
Link, watch, and live — built for continuous transcription
Three signature capabilities for real-time capture, automation, and link-to-text.
Link Transcription
Paste a link, download the media, and generate clean transcripts with history.
Folder Watch
Monitor folders and auto-transcribe new audio/video files as they arrive.
Real-time Transcription
Capture live speech from microphone and application audio with low latency.
From link to transcript in four steps
A fast flow built for link transcription.
Paste a link
We detect the source and extract media info.
Download media
Fetch audio/video locally with retry support.
Transcribe
Whisper runs on-device for accurate text.
Export & share
Export SRT/VTT or keep a searchable history.
Why should you use Audio Note?
Audio Note leverages the open-source Whisper model to provide efficient and accurate speech recognition. In addition, it implements out-of-the-box GPU acceleration, which significantly improves processing speed and efficiency, and provides users with a smooth user experience.
Transcribe Media Files
Supports real-time transcription of multiple audio and video files to text, supporting MP3, WAV, FLAC, ACC, M4A, MKV, etc.
Real-time Transcription Microphone
Listen to microphone audio and transcribe it to text in real time, and display it in lyric mode.
Real-time transcription application
Support real-time transcription of screen and application audio, transnational meetings, online classes, watching live broadcast and other scenarios.
Export Captions
The transcribed text can be exported to a variety of subtitle formats, supporting SRT, VTT, SUB, ASS, SSA, LRC, SBV, SMI, etc.
Record Audio
You can choose a microphone or any input device to record audio and then transcribe it.
Data Security
All transcription is done on your device, with no data leaving your machine, making it ideal for sensitive audio, such as interviews.
GPU Acceleration
Gpu-accelerated transcription is supported on Mac devices with M series chips, and Cuda and Vulkan engines are supported on Windows platforms.
Fallback CPU
Devices without a graphics card can be used to fallback to using the CPU run model to transcribe.
Advanced Settings
Support for adjusting the parameters of the whisper running model (prompt, offset, greedy/beam search, entropy threshold, etc.)
Translation
Support for translating transcribed text.
Record Application
You can choose an app to record and transcribe it later
AI Chat
Speed up your workflow with AI
Frequently Asked Questions
Here are some of the most frequently asked questions.
Still have questions? We're here to help!
Ready to try-out AudioNote?
Quickly convert your audio to text
