Free GPU Transcription
Every session is transcribed with GPU‑accelerated speech‑to‑text and automatic speaker separation. You get accurate, speaker‑labeled transcripts with synced audio playback at no cost, for every session, forever.
Free GPU‑powered transcription for your D&D sessions.
Recording, transcription, playback, search, and sharing are completely free. AI features use a simple pay‑as‑you‑go top‑up at cost. No subscriptions, no tiers, no limits.
Every session is transcribed with GPU‑accelerated speech‑to‑text and automatic speaker separation. You get accurate, speaker‑labeled transcripts with synced audio playback at no cost, for every session, forever.
Add the bot to your Discord server, link it to your campaign party, and record directly from voice chat. Each speaker gets their own audio track captured in real time. When you stop, transcription starts automatically.
Click any line in the transcript to jump to that exact moment in the audio. Search by keyword or speaker, switch between individual speaker tracks or the full mix, and skip over silence automatically.
Generate session recaps and campaign notes with the latest AI models, including Claude Opus 4.6, GPT‑5.2, and Gemini 3 Pro. Bardlog also auto‑extracts a campaign wiki of NPCs, locations, items, factions, and lore that builds across every session.
Select any transcript range and export a styled video or audio clip. Choose from clean transcript overlays, animated styles, or the "brainrot" mode that composites your session over gameplay footage. Built for sharing on social media.
Link D&D Beyond character sheets to your party members. Portraits, character names, ability scores, and equipment appear right alongside your transcripts. Character data is snapshotted for each session.
Organize players into campaign parties, share invite links, and track sessions together. Add public or private timestamped notes to any moment in a session. Link your Discord server for seamless recording.
Not using Discord? Upload WAV, FLAC, MP3, M4A, OGG, Opus, or WMA files directly. Send a single recording or a ZIP with one file per speaker for multi‑track results. Large files are chunked automatically.