MacWhisper 8 debuts with new video player, WhisperKit models and ChatGPT 4o support

MacWhisper is a very useful utility for quickly transcribing spoken audio on a Mac. Version 8 is launching today with new features in video, speed, AI, etc.

Video

The new video player is a flagship feature of MacWhisper 8. It greatly simplifies the process of transcribing audio from video files.

Transcribed audio appears as subtitles. , and the translated text is displayed as separate subtitles. The video player appears embedded by default, but you can turn it into a separate window if you wish.

Speaking of video, MacWhisper 8 also brings improvements to YouTube. The downloaded video can be played in a mini-player, the download can be high or low quality, and downloading from YouTube is usually faster. You can also choose between downloading videos or audio only from YouTube.

WhisperKit

MacWhisper 8 also introduces support for the WhisperKit model.

WhisperKit is the WhisperKit engine that provides real-time speech recognition on an Apple device. WhisperKit models take advantage of hardware acceleration, including the Neural Engine and Apple's CoreML platform, making them especially efficient.

WhisperKit models can be added to the list of models in Settings > Advanced > Show WhisperKit Models .

AI and improvements

MacWhisper 8 finally introduces support for the latest ChatGPT models from OpenAI. This includes GPT-4 Turbo and the recently released GPT-4o, which premiered a week ago.

Meanwhile, MacWhisper includes a lot of general improvements to the app, which you can read about below:

If you have a character limit set, the app will not cut off words in the middle of a word.
New icon in the menu bar that does not conflict with the standard microphone icon.
Quality and language toggles moved to the toolbar. Maximize the window if they are not visible.
You can now open .whisper files while loading models.
Updated to the latest version of the Whisper C++ engine, now with the Flash Attention feature (activate in the menu Settings > Advanced.
Changed the Manage Models screen to make it easier to select a model. Feedback is welcome.
Improved error handling when loading models.
“MS Teams Virtual Mic” excluded from microphone options because it is not a real microphone.
Fixed an issue where invalid license error codes were not displayed.
Fixed a crash when non-pro users added more than two speakers.
Esc key does not close screens during active processes such as recording or batch transcription.
The cloud transcription feature now only displays a list of languages that are supported (57) compared to the 100 that are supported locally.
Bug fixed which caused m4a/mp4 files to be rejected even though they are supported.
All file formats supported Local transcription mode support is now supported for Cloud.
Shortcut key modifiers are now displayed in the user interface (⌘+R, etc.).
Improved button design.
Fixed errors in copying transcripts.

MacWhisper is the main app on my Mac and is a must have if you are interested in artificial intelligence. Get MacWhisper 8 today and marvel at the power of AI-powered transcription on your Mac.