DEV Community

Takashi Abe
Takashi Abe

Posted on

Ultimate Formation to Never Stop Thinking! Starting a Fully Automated Voice Input Life with MacWhisper Pro and AquaVoice

Recently, voice input has been at the center of my life.

Don't you sometimes feel that the act of typing on a keyboard can block the flow of your thoughts?

In light of this, I introduced a new app.

It's MacWhisper Pro (available as Whisper Transcription on the App Store). This time, I want to talk about how I use it alongside my beloved AquaVoice and why voice input using a local LLM is the best option right now.


🎙️ MacWhisper

MacWhisper - Quickly and easily transcribe audio files into text


The Two-Pronged Approach of Real-Time AquaVoice and Batch Processing MacWhisper Is the Ultimate Solution

To get straight to the point, the right answer for me now is to use different voice input environments for different purposes.

When I want to turn my thoughts into text in this very moment, I use AquaVoice. When I want to capture thoughts during reading or long thinking sessions, I use MacWhisper. By combining these two, I can now keep my output running at full capacity at all times.

Why Did I Add MacWhisper?

AquaVoice is currently the best voice input app available for Mac in terms of both accuracy and real-time performance. However, it had some weaknesses.

  • Recording time limitations: Not suitable for long continuous recordings.
  • Manual intervention required: For long sessions, you need to periodically stop and restart, and each time keyboard operation is required, which interrupts your thinking.

What I needed was to transcribe recordings that stream continuously for 30 minutes to an hour while reading—and MacWhisper answered that need.

Local Processing and External Integration Eliminate Friction in Intellectual Production

Why is MacWhisper so easy to use? The reason lies in its design that does not interfere with the user's thinking.

  • One-time purchase option available: Instead of going through cloud APIs, it uses the local LLM (Whisper model) on your Mac itself, making a one-time purchase possible with reduced running costs.
  • Seamless app integration: You can summarize transcribed data with LLM or send it directly to external apps like Obsidian. The fact that it is designed to handle post-writing processing is a nice touch.
  • Privacy and speed: Since you don't need to send data to the cloud each time, you are freed from the waiting time and security concerns typical of web services.

My Personal Voice Input Workflow

Here is how I actually use these tools:

Situation Tool Reason
Thinking and writing in front of my Mac AquaVoice Real-time input provides the best experience
Thinking while reading or during long thinking sessions MacWhisper Can handle long recordings, batch transcription
Taking notes outdoors notebooks Noisy environments make desktop-level accuracy difficult

Failure Story:
What about voice memos while running? I thought it would be great to take notes while running, so I tried it, but it was a failure. Wind noise and car noise significantly reduced accuracy. When outside, the better options for now are bringing the iPhone close to your mouth or using Siri on Apple Watch.

TIPS tested by entering in quote format

Voice Input Is a Thought Accelerator

After trying various things, voice input is just right for us older folks. Above all, it's fast, and the biggest advantage is that your thinking doesn't easily stop.

A tool that lets you convey what you think, exactly as you think it, immediately.

Of course, when organizing and systematizing things, I still use analog notebooks and pens. What matters is distinguishing between outputting (voice) and organizing (handwriting), don't you think?

MacWhisper has a universal environment available on iPhone and iPad as well. A system that catches your thoughts without missing any, whether at home or outside. Why not try building one yourself?

Top comments (0)