Resound 0.1.0 is the first official release of the local-first,
vendor-neutral Discord voice transcription toolkit.
Highlights
- Capture macOS system audio and microphone input with recorder preflight and
device diagnostics. - Control recording through Discord slash commands in mock, local-capture,
Discord-native, or automatic modes. - Transcribe with local Whisper or OpenAI-compatible providers.
- Preserve per-speaker tracks through the experimental Pycord Discord receiver.
- Export portable Markdown, JSONL, VTT, and SRT sessions with summaries and
action items. - Send completed sessions to filesystem, stdout, webhook, Strata, or
TotalRecall sinks. - Validate consent, manifests, output completeness, and Kujo checks.
Security
- Updates PostCSS past GHSA-6g55-p6wh-862q.
- Updates Undici past GHSA-8xcm-r25x-g524, GHSA-m8rv-5g2x-5cg5, and
GHSA-v3r7-h72x-cjcm.
Discord-native voice receive remains experimental pending broader live,
multi-speaker DAVE/E2EE acceptance testing. Local capture and file
transcription are the recommended production paths for this release.
See the changelog
for the complete release summary.