Short answer: Blackboard has no built-in dictation, but you do not need one. A system-level voice typing app types at your cursor in any browser field, so you click into a feedback box, rubric comment, announcement, or discussion reply and speak. Nothing gets installed in your institution's LMS.
Ask any instructor what part of the term they dread and you will not hear "lecturing." You will hear grading. Specifically, you will hear the part of grading where the assessment is already finished in your head after ninety seconds of reading, and then you spend six minutes typing an explanation of it into a small box, forty times in a row.
That gap is the whole problem. The evaluation is fast. The transcription of the evaluation is slow. And because the transcription is slow, feedback shrinks. "Good work, see rubric." "Needs more support." The comment that would have actually changed the next paper never gets typed, not because the instructor did not think it, but because typing it was the forty-first time that afternoon.
Voice dictation closes that gap directly, because the bottleneck was never the thinking.
Does Blackboard have dictation?
Not in the way instructors hope. Blackboard's text editors are standard rich-text fields in a web page, and there is no microphone button waiting in the grading interface for you to click.
Blackboard does support audio feedback in some grading contexts, where you record a clip that the student plays back. That is a genuinely different thing, and worth being clear about. An audio clip is not searchable, not skimmable, cannot be pasted into a rubric row, cannot be translated by a student who needs it in another language, and cannot be scanned by a screen reader without a separate captioning step. It is also awkward for the student who wants to check one specific point while revising at eleven at night.
Dictated text is text. It ends up in the same box a typed comment would, formatted the same way, exported the same way, readable the same way. You get the speed of talking and the artifact of writing.
Which means what you actually want is not a Blackboard feature. It is a system-level voice typing layer that works in the browser you are already grading in.
The part that matters for institutional software
This is the section that makes the difference for anyone working inside a university LMS, so it is worth stating plainly.
Nothing gets installed into Blackboard. There is no LTI tool to request, no building block for your administrator to vet, no integration for the LMS committee to review, no browser extension for IT to whitelist across a campus. Voice Keyboard Pro runs on your Mac, in your menu bar, and types wherever your cursor happens to be. From Blackboard's point of view, text simply arrives in the field, indistinguishable from typing.
Anyone who has ever tried to get a tool approved for a campus LMS knows how much that matters. The approval process for an LMS integration is measured in semesters. This is measured in a download.
Getting set up
On a Mac
Install Voice Keyboard Pro, choose a hotkey, and the workflow is: click into the field, hold the hotkey, speak, release, and your words appear at the cursor.
Two setup notes specific to Blackboard.
Click into the field first. Text goes wherever the cursor is. If you speak before clicking into the feedback box, the text has nowhere to land. This sounds obvious and it is the single most common first-day mistake.
Watch for hotkey collisions. Blackboard's editors and your browser both claim keyboard shortcuts, and grading interfaces are dense with them. Pick a hotkey combination that nothing else is using, and test it once inside the grader before you rely on it during a stack of forty submissions.
A third habit worth building: Blackboard sessions time out, and a long feedback comment lost to a silent logout is a genuinely demoralizing way to lose ten minutes. Submit comments as you go rather than composing five at once, or draft long items in a scratch document and paste them in.
On an iPhone or iPad
Blackboard's mobile apps and mobile site are ordinary text-entry surfaces, so the Voice Keyboard Pro keyboard works there. Switch to it with the globe key and use the mic button built into the keyboard.
Realistically, phone grading is not what most instructors do. Where mobile earns its place is the in-between moments: dictating an announcement while walking across campus, replying to a discussion thread between meetings, capturing a note about a student's question before it evaporates. Voice Edit helps here specifically, because it lets you speak a correction instead of trying to place a cursor between two words with a thumb.
The five surfaces worth dictating
Instructors think of "grading" as one activity. In Blackboard it is really five different writing tasks, and they benefit unequally.
1. Assignment feedback comments
The headline use. This is where the six-minutes-per-student cost lives, and it is where speaking pays back hardest.
A structure that works well spoken, because it is the order you already think in:
- What worked. One specific thing, named. Not "good job" but the actual move the student made.
- The main issue. One, not six. The one that most limits the grade.
- Evidence from the work. Where in the submission you saw it.
- What to do differently. Concrete, actionable on the next assignment.
- The forward look. One sentence on what you want to see next time.
Spoken, that is about forty-five seconds. Typed, it is the six minutes that made you write "see rubric" instead. The structure matters more than the speed, because it is what stops dictated feedback from wandering.
2. Rubric criterion boxes
The most neglected high-value field in any LMS. Every rubric row has a small comment box, and almost nobody fills them in, because filling in six small boxes per student is the most tedious typing in academia.
Those boxes are also the most useful feedback the student receives, because they attach the comment directly to the criterion. One sentence per row, spoken, takes a few seconds each. This single habit changes what a rubric is worth.
Type the scores. Speak the sentences. More on that boundary below.
3. Announcements
Course announcements are where tone gets lost. Typed under time pressure, they come out clipped, and students read clipped as annoyed. Spoken, they carry the register you would use in the room: here is what changed, here is why, here is what you need to do, here is when.
Dictate it, read it back once, and post. The read-back matters because speech runs longer than writing and announcements should be short.
4. Discussion board replies
Discussion boards are graded on instructor presence, and instructor presence dies when replying is expensive. Dictating a substantive reply takes under a minute, which is the difference between replying to four threads and replying to fourteen.
One caution. Discussion replies are public to the class, and speech is looser than typing. Read before you post, particularly if the thread is contentious.
5. Assignment instructions and course content
The writing you do before term starts, when nobody is watching and everything is late. Assignment prompts, module overviews, page descriptions, item text. This is ordinary long-form prose and it dictates as easily as any other prose.
One useful trick: talk the assignment prompt through as if explaining it to a student in office hours, then cut it down. The version that comes out of that pass is almost always clearer than the version written in formal prose from a blank field, because you naturally anticipate the questions.
Speak the sentences, type the values
This is the boundary that keeps voice from causing problems in an LMS, and it is not optional.
Dictate: feedback prose, rubric comments, announcements, discussion replies, assignment instructions, emails to students, notes to yourself.
Type by hand, every time:
- Point values and scores
- Percentages and weights
- Due dates and time windows
- Student IDs and course reference numbers
- Rubric level selections, dropdowns, and radio buttons
- Anything in a grade column
The reason is that number errors are silent. A garbled word in a comment is visibly wrong and you fix it. A score of 18 that lands as 80 reads as a completely normal number and travels straight into the gradebook. In grading, silent errors are the expensive kind, because the student who was under-credited may never say anything.
If you want the mechanics of how spoken digits get interpreted, we covered that in detail in how to dictate numbers and dates. But for grading, the safe rule is simply that no figure that affects a grade should ever be spoken.
The name problem
Here is where stock dictation quietly fails instructors. Every transcription system guesses at words based on how common they are in general speech, and a class roster is full of names that are not common in general speech. So is your discipline's vocabulary.
Student names come back wrong. Course codes come back wrong. Technical terms specific to your field come back wrong. And they come back wrong the same way every time, because stock dictation does not learn from your corrections. You will fix the same student's name in twelve consecutive comments.
Smart Vocabulary in Voice Keyboard Pro is a personal dictionary with replacement rules, so you load the terms once and they come out right afterward. For an instructor, a sensible starter list is your current roster, your course codes and section identifiers, the ten or fifteen discipline terms you use constantly, your textbook and author names, and your colleagues' names.
A curation note worth stating: if you would rather not put full student names into a local dictionary, initials or first names work fine, and the payoff is nearly the same. That is your call and your institution's policy, and it is worth thinking about for thirty seconds before you load a roster.
Feedback quality also improves for a reason that has nothing to do with speed. When you are typing, you avoid words that are annoying to type, including students' names. Dictated feedback naturally addresses the student by name more often, and named feedback reads as feedback rather than as a form letter.
How to keep dictated feedback defensible
Grading comments are records. They get appealed, they get forwarded to chairs, and occasionally they get read by people you did not write them for. Speech is fluent, and fluency makes people say more than they meant to.
Describe the work, not the student. "This argument does not have supporting evidence" rather than "you did not try." The first is an assessment. The second is a characterization.
Observations, not diagnoses. Note what is in the submission. Do not speculate about why, about the student's circumstances, or about anything you cannot see in the work.
Never submit a hot comment in the same sitting. If a submission irritated you, dictate the comment, then move to the next one and come back. Speech carries irritation more directly than typing does, because typing gives you the fifteen seconds of friction that used to cool you down. That friction is gone now, and you have to supply it deliberately.
Read before you submit. Every time. It takes ten seconds and it catches the sentence that came out sharper than you meant.
Your institution's academic policy and its rules about what belongs in a student record are your registrar's and your department's call, not something a software blog can rule on. If you are unsure whether a comment belongs in the LMS, that is a question worth asking before you post it, not after.
A grading session that actually works
Putting it together for a stack of forty submissions:
- Read three submissions before writing anything. This calibrates you and stops the first student from getting your harshest or softest comments.
- For each submission, read fully, then speak the five-beat structure in one pass. Do not stop to fix wording mid-comment.
- Speak the rubric criterion boxes, one sentence each. This is the habit that changes the most.
- Type the scores. Always by hand, always looking at them.
- Read the comment once. Submit.
- Take a break every ten. Voice fatigue is real, and rushed talking is less accurate talking.
That last point deserves a mention, because instructors moving to voice for the first time often try to grade forty in one unbroken session and end up hoarse. Sip water, pause between batches. We wrote about that pattern more fully in voice strain from dictation.
Multilingual courses and international students
If you teach students who are working in a second language, dictated feedback has a practical advantage: it is longer and more explanatory than typed feedback, which is exactly what a student reading in a second language needs. Terse feedback is hardest on the students who most need clarity.
On the iPhone side, Voice Keyboard Pro includes two-way translation while dictating across 24 languages, which is useful for short communications with students or families who are more comfortable in another language. That is a communication aid, not a substitute for your institution's official translation and accessibility services, and anything formal should go through those channels.
Frequently asked questions
Does this work with Blackboard Ultra and Original?
Yes to both, and to the mobile site. The approach does not depend on Blackboard's interface at all. It types into whatever text field has focus, so course view, grading interface, and content editor all behave the same.
Does my IT department need to approve it?
Nothing is installed into Blackboard or into your institution's environment, so there is no LMS integration to approve. Whether your institution has rules about software on a managed university-owned Mac is a separate question, and that one is genuinely your IT department's call. On a personal machine it is a normal app install.
Is this the same as Blackboard's audio feedback?
No. Audio feedback produces a recording the student listens to. Dictation produces text in the comment box. Text is skimmable, searchable, translatable, and accessible to screen readers without extra steps, which is why most instructors who try both end up preferring dictated text for anything substantive.
What about student privacy?
Voice Keyboard Pro's server stores only operational pings. No audio and no transcript content is stored, so the feedback you dictate stays local to your machine. That said, whether student work and student data may be processed by any given tool is a policy question your institution owns, and if you handle records covered by student privacy regulations, ask your privacy office rather than relying on a vendor's blog post.
Can I dictate in other LMS platforms too?
Yes, and the workflow is identical, because none of it is Blackboard-specific. We have written the same guide for Canvas and for Google Classroom, and the surfaces map almost one to one.
Will it handle discipline-specific terminology?
Not out of the box, because your field's terms are rare in general speech. Load them into Smart Vocabulary once and the problem goes away. If you write in LaTeX or work with heavy notation, dictating math symbols and equations covers what voice can and cannot do with notation, and the honest answer is that characters work and two-dimensional structure does not.
What this actually changes
The pitch is not that you will grade faster, though you will. It is that the feedback students receive will be longer, more specific, and more often addressed to them by name, because the thing that was making it short was never a lack of things to say.
An adult types at roughly 40 words per minute and speaks at 130 to 150. In a job where the deliverable is explanatory prose produced forty times in an afternoon, that ratio is the entire argument. The rubric boxes get filled in. The announcement gets the sentence explaining why. The discussion thread gets a real reply.
If you want the broader picture of voice in teaching work, voice to text for teachers and voice typing for professors cover the rest of the week. Voice Keyboard Pro has a free tier with daily limits, which is enough to test it on one grading session before you decide. Pro is $4.99 a month or $34.99 a year.