Upload the recording, or skip straight to text.
Start a session in your campaign and give Everlore whatever you already have. Most tables upload the recording from game night. Tables that already keep transcripts can skip transcription entirely.
- One mixed recording of the whole table.
- Separate tracks, one per player, from a voice-chat recorder.
- An existing transcript or written notes, pasted or uploaded as text.
- Common audio formats including MP3, WAV, M4A, FLAC, OGG, Opus, and WebM.
Separate tracks give the best result, because each player is already isolated and speakers are identified by track instead of guessed from a shared microphone. They also cost more managed transcription time, since every track is metered separately. Companion transcribes multitrack locally without touching your allowance.
3 h 42 m · one mixed recording · speakers detected during transcription
Each speaker already isolated, so names attach to the right voice. Metered per track.
Already have the words? Everlore begins at analysis.