Short answer: it will work, partially, and the partial part is sneaky. I build a competing tool, so read me with appropriate suspicion, but I’ll make the case the way I’d make it to a friend: here’s exactly what Otter does well at your table, here’s where it quietly hands the job back to you, and here’s how to tell which kind of group you’re in before you’ve burned a month finding out.

What Otter genuinely does well

Otter is a good transcriber. Feed it clear audio and you get clean, timestamped, speaker-separated text at conversation speed, with a summary on top. It’s inexpensive, the editor is pleasant, and for its actual job, meetings, it might be the best-known tool on earth. None of that evaporates because your meeting has a dragon in it. If what you want is a searchable text record of one session, Otter will hand you one, and the first time you watch your four-hour game turn into scrollable text, it feels like the future.

So if your table’s whole ambition is “we’d like transcripts,” you can stop reading and go set it up. Genuinely. A table with transcripts is doing better than ninety percent of tables.

The gap shows up later, and it has three parts.

Gap one: speakers are people, but the game is characters

Otter will correctly figure out that six humans spoke, and with a little training it will even label them by name. Here’s the problem: Mike is three entities. He’s Mike when he asks whose turn it is to get the pizza, he’s Kalden the ranger when he’s in character, and he’s half the NPCs in the scene when the DM is busy. A meeting transcriber has no slot for that. “Mike: I draw my sword” and “Mike: the innkeeper draws HIS sword” look identical to it, because in a meeting they would be the same kind of sentence.

That mapping, speaker to character to who’s-actually-talking-in-the-fiction, is the single most load-bearing piece of context in a session transcript, and no general-purpose tool extracts it, because it only exists at gaming tables. You can reconstruct it while reading, sure. You, personally, every time, forever.

Gap two: fantasy names eat transcription vocabularies alive

Transcription models are trained on how people talk, and people mostly do not say “Szeth-son-son-Vallano” or “the Duchy of Virelosse.” Your proper nouns will come back creatively mangled: Vex becomes “Beck’s,” Zyrathis becomes “zero of this,” and the Duchy of Virelosse becomes a fresh surprise every single week. Otter lets you teach it custom vocabulary, which helps, and which also means you’re maintaining a glossary of your campaign’s proper nouns by hand, updated every time the DM invents a name mid-sentence, which the DM does constantly.

We hit the same wall at Loracle, because every speech model hits it. Our answer was to stop fighting it at the audio layer and fix it at the memory layer: when the campaign’s own record already knows there’s an NPC called Zyrathis, a phonetically similar mangle in the next transcript gets reconciled back to the real name instead of founding a new character. The fix needs a campaign memory to reconcile against. A tool that sees each file once has nowhere to put one.

Gap three: every session is an island

This is the big one, and it’s structural rather than a missing feature. Otter’s unit of the world is the conversation. Your campaign’s unit is the arc: the favor owed in session three that detonates in session eleven, the fisherman who turns out to be the duke, the prophecy whose exact wording suddenly matters eight weeks after it was spoken. Fifteen clean transcripts in a folder answer none of that, because the connections live between the files, and between the files is exactly where a per-conversation tool never looks.

So Wednesday night arrives, your rogue asks whether the smuggler debt is still live, and your options are: search fifteen transcripts for “smuggler” and read every hit in context, or do what tables have always done and argue from memory. I’ve written about what that gap between sessions does to a group, and about why the Craig-plus-elbow-grease pipeline breaks down at the same seam. The pattern repeats with every general tool: the recording layer is solved, the single-transcript layer is solved, and the campaign layer is always, politely, your problem.

The honest answer

Yes, you can use Otter for D&D, and here’s the profile of the table it works for: you want transcripts more than you want answers, someone at your table actually enjoys being the archivist, and your campaign is short or episodic enough that cross-session questions are rare. One-shots, short arcs, a podcast table that edits anyway. For that table, Otter is a fine tool and cheaper than anything purpose-built.

The table it fails is the one running a long campaign, which is most tables that get this far into a post about transcription. For them the archivist job compounds weekly: export, upload, fix the names, re-read old transcripts to answer new questions, maintain the glossary. Every step is small. The sum is a part-time job that lands on the one person who already DMs.

Loracle’s bet is that the campaign layer is the actual product. It sits in your Discord call, you type /record, and after the session the transcript arrives already speaker-labeled, in whatever language your table plays in, with the recap posted to your channel. Then the part Otter structurally can’t do: every session feeds a living record of the campaign, NPCs and debts and prophecies connected across every session they touch, so when the smuggler argument starts, anyone types /chat and gets the answer from the actual scene, with the wording intact.

Try the Otter route first if you’re unsure; it’s the cheapest way to discover which table you are. Just set a reminder for six weeks out that asks one question: “am I still doing the pipeline by hand?” If the answer is yes and you’re still happy, you’ve found your tool. If the answer is yes and you’re tired, the problem was never the transcription.

And it works. Let me say that clearly before I say anything else, because I build a paid tool that lives one shelf over from this workflow and you should know my angle. Paste one session into a chatbot and ask for a recap, and you will get a decent recap. Ask it who took the deal with the smuggler, and it will usually tell you. For one session, in one window, on one Sunday, the free DIY version genuinely delivers. If your campaign were one session long, I’d have nothing to sell you.

Campaigns are long. That’s the entire plot of this post. Three things go wrong between week one and week six, they go wrong quietly, and each one lands on the same person: you.

Week one: the recap is good, and you did four jobs to get it

Count the steps you just did by hand. You exported the audio. You ran it through a transcriber. You read enough of the output to fix the worst damage, because the transcript came back calling your warlock’s patron “Sir Rathis” in one scene and “zero of this” in another. Then you pasted it, waited, read the summary, and fixed the two places where it attributed the ranger’s kill to the bard.

None of those steps is hard. That’s exactly what makes the trap comfortable. Each step is ten minutes, the whole pipeline is under an hour, and the result really is useful, so you do it again next week. You’ve volunteered for a recurring shift, and like most recurring shifts, it was easiest to accept in the week you had the most enthusiasm. Week one is always the week you have the most enthusiasm.

The names drift, and the chatbot papers over the drift

Here’s the failure that surprises people, because it hides inside a strength. Chat models are agreeable readers. Hand one a transcript where the lich’s name arrives as “Vel Karoth,” “Velkorath,” and “the whole karath,” and it will smooth all three into one villain without being asked. Convenient, except it has to guess which spelling is canon, and it guesses fresh every week, because next Sunday is a new chat window with no memory of last Sunday’s guess.

So session four’s recap stars Velkorath and session seven’s stars Vel Karoth, and six weeks later your notes doc contains two liches, or one lich under a name nobody at the table actually says. There’s no glossary anywhere in this pipeline. The model in the window can’t keep one, the transcriber doesn’t know your campaign exists, and the one place the canonical spelling lives is your head. Every proper noun the DM invents mid-sentence, and the DM invents them constantly, gets re-adjudicated weekly by a reader who wasn’t there.

We fight this exact problem at Loracle and I can tell you from the inside: the fix lives outside the transcript. You check each new mangle against a persistent record of the campaign, where Vel Karoth already exists as an entity with a spelling, so the next phonetic casualty gets reconciled to him before it can found a rival lich. The fix is memory. A chat window is the one place in this workflow that’s guaranteed to have none.

Every window is session one

Which brings us to the structural problem, the one that no amount of careful pasting gets you past. A chat conversation holds what you put in it. Your campaign’s best material lives between the sessions: the favor from session three that comes due in session eleven, the fisherman who’s been the duke since session two, the exact wording of the prophecy from the night half the table was giddy on cough syrup.

Your rogue’s Wednesday question is almost never about last session. It’s “do we still owe the smuggler, or did the barge thing cancel it,” and the answer is split across three transcripts in a folder. Now your options are: start a fresh chat, paste in all three (assuming you can tell which three), and hope the answer survives a context window that fits four hours of table talk the way an overhead bin fits a cello. Or do what every table without a record does: argue from memory and let the loudest version win.

Some of you will now invent the obvious patch, because you’re the kind of person who’s read this far: keep a running “campaign so far” document, paste that in alongside each new transcript. It helps. It’s also a fifth manual job, it grows every week, and you’ve just rebuilt a campaign wiki with yourself as the sync mechanism. At that point you should read what a campaign wiki is actually for, because you’re building one on hard mode.

The honest answer

Yes, you can paste the transcript into ChatGPT, and for some tables that’s the right call. Here’s the profile: your campaign is short or episodic, your questions are mostly “recap last session” rather than “trace this thread across the fall,” and someone at the table genuinely likes being the archivist. One-shots, six-session arcs, a West Marches table where every night is self-contained. For that table, DIY is free and fine, and I mean both words.

The table it quietly fails is the long campaign, where the questions compound and the pipeline runs every single week whether you’re tired or not. The money cost stays zero. The cost is an hour of your Sunday, a glossary that lives in your head, and a stack of summaries that can’t talk to each other. You weren’t quoted that price up front, which is the only dishonest thing about the whole workflow.

Loracle is the version of this where the pipeline is nobody’s job. It sits in your Discord call, you type /record, and after the session the transcript shows up speaker-labeled in whatever language your table plays in, the recap posts itself to your channel, and every session feeds one connected record of the campaign, the NPCs and debts and prophecies linked across every session they touch. When the smuggler argument starts, anyone types /chat and the answer comes back from the actual scene.

If you’re unsure which table you are, run the DIY pipeline for a month; it’s the cheapest possible experiment. Then notice which Sunday you start putting it off. That Sunday is your answer arriving.