Why I love MacWhisper

Over the past few months, I have experimented with various generative AI and AI technologies, protocols, and platforms. One application that has caught my attention is MacWhisper, developed by independent developer Jordi Bruin. MacWhisper utilizes OpenAI’s Whisper technology for accurate text transcription.

Whisper, created by OpenAI, is a powerful tool that converts spoken words from different sources, such as phone calls or videos, into written text. It seamlessly works with various apps and allows users to control them through voice commands. Continual improvements are made to enhance its ability to connect individuals with computers in the ever-changing digital landscape. OpenAI developed Whisper by scraping YouTube and openly available podcasts and audio content to extract information and refine its models.

Compared to its more limited competitors, Whisper is trained to outperform them in various tasks, such as taking commands or converting voice to text. For example, while Apple’s Siri on the iPhone is often mediocre or downright terrible when using ChatGPT’s mobile app, it effectively transcribes voice into text.

One of the key advantages of MacWhisper, as it utilizes the Whisper technology, is its superior accuracy in transcriptions compared to other services I have used. I was an early adopter of Rev.com and eventually switched to Otter.ai, for which I paid approximately $100/year. However, I have since canceled my subscription.

Don’t get me wrong—Otter is excellent for automatic transcriptions of online meetings using platforms like Zoom and Google Meet. It maintains a comprehensive record of all recordings and transactions on its platform, offering numerous features. 

Since MacWhisper processes audio files locally, and the developer offered a free basic version, I decided to try it. If you require medium or large transcription models, you upgrade for ten euros or more, depending on your preference and generosity. It supports audio formats such as MP3, WAV, M4A, and MP4 for video files. Simply drag and drop the files onto the app, and you’re ready to go. It’s worth noting that MacWhisper is a resource-heavy application, and it performs best on M1 or M2-based Macs. The first time the fans on my M2-based MacBook Pro turned on was when I transcribed an hour-long interview using MacWhisper.

After comparing the audio transcripts from MacWhisper and Otter, I found that while Otter’s formatting is superior, there is hardly any difference in the transcriptions. I started with the free model, which was sufficient for my needs, and eventually upgraded to the large models. Nowadays, I primarily use MacWhisper to transcribe my voice memos, which I then import into Lex.Page. From there, I give it a command to create a list-like entry for my daily work journal.

Of course, MacWhisper could still be improved. It would be fantastic if it could learn from my voice samples, adapt to my accent, and become more accurate over time. It would also be beneficial if the transcription included suggested improvements, such as correcting spelling mistakes. In an ideal world, Jordi would integrate MacWhisper with other apps, such as his MacGPT, and provide links to topics of interest within the transcription itself.

MacWhisper exemplifies how “AI” can enhance applications and inspires other developers to harness the powerful capabilities of M1 and M2 Macs creatively.

Bottom line: If you are spending $100 bucks on services like Otter, I suggest, you give this a try, at the very least. This is the best 10 euros I have spent in 2023.  

MacWhisper App website.