Why I No Longer Use Dragon in 2026: A Personal Reflection
Voice recognition has been part of my life for decades, long before it became mainstream. My interest in the technology goes right back to childhood, when I watched Star Trek: The Original Series and saw characters casually talking to computers as if it were the most natural thing in the world. I always hoped we’d reach that point — a future where you could simply speak, and the computer would understand.
Today, with modern AI dictation, I feel we are finally almost there. But the journey to this point has been long, expensive, and full of trial and error.
---
My First Computer: The BBC Micro (1986)
My first computer was a BBC Micro 32K, which I got around 1986. It had no mouse, no graphical interface, and certainly no voice recognition. I even remember asking whether any kind of speech system existed for it, but of course nothing like that was available at the time. Still, the dream of talking to a computer stayed with me.
---
My Early Mac‑Only Voice Recognition Tools
Before Apple built high‑quality dictation directly into macOS, the Mac platform relied on a handful of third‑party voice recognition systems. Over the years, I tried many of them — often spending a lot of money on products that didn’t always live up to their promises. But each one was part of the journey.
---
MacSpeech iListen (Released before 2006, discontinued January 2008)
iListen was one of the earliest Mac‑only dictation tools. It used the Philips speech engine, not Dragon’s.
• Released: Early 2000s (in wide use by 2006)
• Discontinued: January 2008, replaced by MacSpeech Dictate
In practice, I found iListen patchy and unreliable. It struggled with accuracy, required very careful pronunciation, and often forced me into constant corrections just to get a usable sentence. It was one of the few native Mac options at the time, but compared to what came later, it felt slow, inconsistent, and far too demanding for everyday use.
---
MacSpeech Dictate (Released February 15, 2008)
When iListen was discontinued, MacSpeech released MacSpeech Dictate on February 15, 2008.
• Engine: Dragon NaturallySpeaking (licensed from Nuance)
• Discontinued: After Nuance acquired MacSpeech in 2010
This was a major improvement because it finally brought Dragon’s accuracy to the Mac. It was the first time dictation on macOS felt genuinely usable.
---
IBM ViaVoice for Mac (Released early 2000s, discontinued mid‑2000s)
ViaVoice was one of the first major dictation systems available for Mac OS X.
• Released: Early 2000s
• Discontinued: Mid‑2000s, after IBM/ScanSoft/Nuance stopped Mac development
I tried ViaVoice as well, but I never found it particularly reliable or accurate. Whether it was the software itself or simply the way I had to use it, I often ended up doing multiple corrections just to get a clean sentence. Compared to today’s tools, ViaVoice felt slow, inconsistent, and far too demanding for everyday work.
---
Dragon Dictate and NaturallySpeaking: A Major Leap Forward
My serious journey with voice recognition began in the mid‑1990s with Dragon Dictate 1.0, followed by Dragon NaturallySpeaking. These were the first systems that allowed continuous speech, and they represented a huge leap forward. Over the years, I upgraded through many versions, eventually reaching Dragon 15 and later Dragon Professional 16 on Windows.
Dragon was powerful, but it still belonged to an era where the user had to adapt to the software. It required:
• Microphone calibration
• Profile training
• Spoken correction commands
• Careful setup
• A Windows environment
For a long time, it was the best option available — but it demanded a lot of effort.
---
My Physical Setup Matters
Because of my cerebral palsy, the physical effort required to manage equipment is a major factor in what technology I can realistically use. Dragon expects a dedicated microphone, ideally a freestanding one placed at the correct distance and angle.
I’ve considered buying a freestanding mic, but that would mean:
• Additional cost
• Finding a stable place on my desk or on top of my PC tower
• Adjusting and positioning it regularly
• Managing cables and hardware I can’t easily reach
For me, that setup simply isn’t practical.
By contrast, my iMac’s built‑in microphone is:
• Always in the right place
• Completely hands‑free
• Consistent and reliable
• Ready the moment I start speaking
That convenience alone makes a huge difference in my daily comfort and independence.
---
Windows vs macOS: A Clear Preference
Dragon 16 only runs on Windows, and I’ve found that Windows — especially Windows 11 — is not to my taste. It requires more steps, more menus, and more physical effort to navigate. My iMac, on the other hand, is:
• Faster for me to operate
• More intuitive
• Less cluttered
• Better integrated with the apps I use every day
My workflow lives on the Mac now, and Dragon simply doesn’t fit into that environment.
---
Modern AI Dictation Is Faster and More Natural
Dragon is still powerful, but it comes from an era where the user had to adapt to the software. Modern tools adapt to me.
Today I use:
• Apple’s built‑in voice recognition for quick, instant dictation
• MacWhisper, which uses the Whisper AI model for long, accurate transcription
These tools give me:
• No microphone setup
• No profile training
• No calibration
• No Windows overhead
• Faster results
• Better accuracy in real‑world conditions
For my needs, they simply work better.
---
Older Voice Recognition Required Command‑Based Corrections
One of the biggest differences between older systems like Dragon NaturallySpeaking and modern AI dictation is how corrections are handled.
Dragon relied heavily on spoken correction commands such as “correct that,” “choose 1,” or “spell that.” In reality, I didn’t use those commands very much. They were slow, fiddly, and required precise timing and phrasing.
Modern voice recognition is far more flexible. I can simply:
• Move the cursor
• Make a quick edit with the keyboard
• Continue dictating naturally
And none of this disrupts the text I’ve already dictated.
This freedom to mix voice and keyboard input makes the whole process smoother, faster, and far less physically demanding.
---
The Bottom Line
Dragon remains a strong product, but it no longer fits my workflow, my physical setup, or my preferred operating system. The combination of macOS, Apple’s built‑in dictation, and MacWhisper gives me:
• Less physical effort
• Faster startup
• Better integration
• Fewer steps
• No extra hardware
• Excellent accuracy
• Easy keyboard interaction without breaking dictation flow
After decades of chasing the dream of natural speech recognition — from the BBC Micro to Star Trek fantasies — I finally feel that we’re close to the future I imagined as a child.
Made in RapidWeaver