Let your computer read your draft back to you

Even after multiple readings, I can still overlook repeated words in my writing. My eyes, while knowing the intended meaning, fill in the gaps. My ears, however, are less tolerant of such errors. When I hear a draft read aloud, issues like clumsy phrasing, missing words, and repeated letters become glaringly apparent. I find that reading the text myself helps, but I often modify the content as I go, which disrupts the flow.

Using a computer to read the draft back to me is a more honest approach. Most systems already include a text-to-speech feature, which can be utilized without any additional cost. First, convert the Markdown file into plain text, eliminating formatting elements like asterisks, links, and heading markers. Tools like Pandoc simplify this process with a single command.

The plain text version retains the words while removing most of the syntax. Links become their text equivalents, and emphasis markers disappear. However, code blocks remain intact for technical posts. Let the computer read the text aloud on macOS using the "say" command, adjusting the reading speed with the "-r" option. On Linux, the espeak-ng tool serves the same purpose, albeit with a more robotic voice.

Its "-s" option allows setting the speed in words per minute. For Windows users, the built-in speech engine in PowerShell can achieve the same result. Alternatively, many editors and browsers offer a "read aloud" function. Regardless of the tool used, what matters most is adopting the habit of listening to your drafts. Create a simple script to automate the process, making it a part of your routine.

As you listen, jot down any glaring errors to address in a single editing pass. This method complements structural editing by catching subtle issues that spell checkers might overlook. Although it adds a few minutes per article, the peace of mind gained from a more polished final product is well worth the effort.