Over the years, I have heard David Sparks (@Macsparky) speak with enthusiasm about the benefits of dictating initial rough drafts to be edited later. I had no particular disagreement with his approach, but having written from keyboards for nearly fifty years, dictation was not my normal workflow.
Recently, I decided to give it a try to see whether it might speed up my book writing. One of my objectives for voice and tone in my nonfiction book is to be more personal and a little less academic. The book is meant to read as a mentor writing to a younger mentee. When typing, because of how much academic writing I have done, my prose tends toward the third person, rather formal, and not particularly warm. I wondered whether dictating might enable me to personalize my prose.
The difference has been striking. When I type, I might write something like: “Effective leaders must develop the capacity to make decisions under conditions of uncertainty.” When I dictate, the same idea comes out: “You are going to have to make hard calls without all the information you wish you had. That is simply part of the job.” The dictated version sounds like me talking to someone I am trying to help.
I have been experimenting with this approach (this post was written using this approach), and I have found it extremely helpful. Not only is it improving the tone and warmth of my writing, but it is proving to be much faster. My plan going forward is to continue using dictation for the first, extremely rough drafts. Then I will clean up the text using AI assistance before editing from the keyboard for my final version. In addition to the benefits mentioned, dictation helps ensure that the words I write are mine, not AI’s: my words (dictation)—>AI edits—>my final edits.
If you have been hesitant to try dictation, I would encourage you to experiment with it for a week. The results may surprise you.
I just use Voice Memos. I see no reason to pay a subscription. Voice Recorder provides a transcript of the recording, which I have cleaned up in Claude.
Sorry, I meant Voice Memos, not Voice Recorder. I do have Voice Recorder (a subscription app I tried), but I’m not using it. My bad. I corrected my reply.
I tried this yesterday. First with Notes (no go, if you lock the phone it stops recording, or did in my case) and then with Voice Memos.
Started it up, put the phone in my pocket and when for a walk. 10 minutes later I had the recording and transcript.
I will have to confess, based on the transcript, my elocution could do with some work! But very happy with the output. Punctuation seems fine and will need a clean up but the words and idea are there.
I’m intrigued, Barret, what stopped you doing this earlier? And what changed that meant you’d give it a go?
(I ask because you’re a very thoughtful and introspective guy, and I study change processes as part of my work, so I’d love to know what you think.)
First, thank you for your very kind comments. Undeserved, but appreciated nevertheless.
Second, in answer to your question, I wish I had a more sophisticated response. The truth is that I want to make faster progress on the large book project. Given my busy schedule, including evening events, this can be difficult. After listening to a recent MPU podcast in which @MacSparky mentioned again the value of dictating, and after rereading his November newsletter, “Why Dictation Is Easier and More Relevant,” I decided to try something different.
Another catalyst is AI, which reduces the friction in processing dictated text. Before AI, one could certainly dictate, but the cleanup reduced some of the productivity gains. It was often more efficient simply to type than to dictate, clean up, and edit. AI eliminates that inefficiency.
So, I changed (or at least I’m experimenting) because David said I should and I want to get text down faster.
For what it’s worth, I read the difference between these two sentences as one sounding more like written language and one more like spoken language:
When I type, I might write something like: “Effective leaders must develop the capacity to make decisions under conditions of uncertainty.” When I dictate, the same idea comes out: “You are going to have to make hard calls without all the information you wish you had. That is simply part of the job.” The dictated version sounds like me talking to someone I am trying to help.
So, yes, more like “talking,” if that is the desired effect.
I’ve done both, on and off, for decades (I had a great PA who managed to capture and make sense of my ramblings better than I could, and I used machine dictation too).
I never found much difference in finding a tone or style however I was writing, but then I always hated the 3rd person, impersonal, passive voice academic style when I was required to produce it and instinctively want to be more personal and direct, though not necessarily informal. Where I do notice a difference is in the “stage” of writing. A first draft is often quicker and easier with dictation - if I have a rough idea of what I want to say I can speak it quickly. When I want to explain or make particular points, and find the best way of expressing them, using a keyboard wins for me. I can think about exactly what I am writing more easily. Journalism, especially when I am aiming for a word count against a deadline, is always straight to keyboard: I’m editing and generating at the same time. I don’t want to ramble and then have to edit it radically.
Dictating to a personal assistant was the best. I could instantly tell (without even needing to glance over) when the meaning was not perfectly clear or the intended style was not working and I’d amend my approach, often without needing to think about it.
Yes, that is the intent, with a caveat. I do not want the final written text to sound spoken, but I do want it to be warmer, more personal, and less academic. By starting with spoken text and then editing, rather than typing and editing, I find that I can achieve warmer written prose more efficiently. I would not publish, “You are going to have to make hard calls without all the information you wish you had. That is simply part of the job.” That is too casual. Instead, I would edit it to something like, “As a leader, you will face decisions that must be made with incomplete information during periods of rapid change. Such moments will require both courage and fortitude.” The result is more personal than the first example and more formal than the second.
The actual 1st draft version became (I’m skipping a lot of text, this is an excerpt):
… That unease is sharpest when you must act before certainty arrives. As a leader, you will face decisions that must be made with incomplete information, during periods of rapid change, and often against stiff opposition. Such moments demand both courage and fortitude … Leadership demands both: the courage to make hard decisions with incomplete information, and the fortitude to hold that course when consequences unfold and pressure mounts. One is the virtue of the moment; the other is the virtue of the long road.
I’ve been using dictation to transcribe text for my two blogs now for four years. I too was inspired by @MacSparky. I didn’t come from a background of doing a lot of dictation. I felt very self-conscious at first and It took me some time to get used to it.
Like you, I found my dictation was much more personable and easier to understand. I like the tone I end up with much better than when I wrote my first draft. Writing tended to sound much more formal.
The real clincher was my wife’s assessment: she said that she enjoyed reading my posts much more after I started dictating the first drafts. I haven’t looked back and am still dictating my first drafts.
I wrote a blog post about my experience and the benefits of dictation here:
I had a few minutes of down time and was able to read your blog article. It was great! I have two questions, if you don’t mind:
Do you usually construct a mind map in MindNode prior to dictating any article of substance?
Tell me about your experience with Whisper Memos. Has it been good? Have you compared it with other apps? Also, is there an iPad version? I do not have my iPad with me to check. I do a fair amount of writing on the iPad, which is the reason for my question.
Thanks for sharing your article and your thoughts!
Unless it’s only a few paragraphs (which is almost never), I always build a mind map in MindNode prior to dictation. I find a mind map works better for my graphic mind than an outline. It helps me to get some basic thoughts down, then expand them or add additional points. It also helps me to see gaps in my thinking. I add quotations I might use that I’ve gotten from research. I’ll return to the mind map multiple times over a couple of days to give it time to incubate. When I feel it’s “ready,” I’ll dictate my rough first draft.
I’ve periodically tried using other popular apps for dictation (the last was Mac Whisper), but always return to Whisper Memos. It’s easy to use, it does a great job dividing text into paragraphs, and also offers a summary of my post with bullet points which I sometimes take a look at when editing. It works on iPhone, iPad, and Mac (as an iOS app). I’m usually in front of my Mac when dictating a blog post, but I’ve also done a lot of dictating on my iPad for journaling. Whisper Memos works best for my desires and needs, but you might find that something else fits you better.
I hope this answers your questions. Let me know if you have additional ones.
I agree. I have typically used outlines in OmniOutliner for larger writing projects, but until recently, I had not made dictation a part of my workflow. I am still experimenting, but thus far, it is working well. Thank you for periodically mentioning the value of dictation. Some of us are slow learners!
and I want to get text down faster.