gortium/infra
1
1
Fork 0
Code Issues 13 Pull Requests 21 Actions Packages Projects 1 Releases Wiki Activity

feat(hermes): add Dockerfile for voice support with GPU STT #25

Closed
Hermes wants to merge 4 commits from feat/hermes-voice-gpu-support into master
pull from: feat/hermes-voice-gpu-support
merge into: gortium:master
Conversation 0 Commits 4 Files Changed 7 +632 -1
Hermes commented 2026-04-30 15:16:09 +00:00
Collaborator
Copy Link

Summary

Adds local Speech-to-Text (STT) capabilities to Hermes Agent by extending the upstream image with audio capture libraries and faster-whisper for GPU-accelerated transcription.

Changes

  1. docker/hermes/Dockerfile - New file extending upstream image with:

    • libportaudio2 for audio capture
    • sounddevice, numpy, faster-whisper via uv
    • Health check for voice dependencies

  2. assets/compose/ai/compose.yml - Updated hermes service:

    • Changed from image: to build: for reproducibility
    • Added ROCm environment variables for AMD GPU

Deployment

cd /opt/data/infra
sudo nixos-rebuild switch --flake .#lazyworkhorse

Related

## Summary Adds local Speech-to-Text (STT) capabilities to Hermes Agent by extending the upstream image with audio capture libraries and faster-whisper for GPU-accelerated transcription. ## Changes 1. **docker/hermes/Dockerfile** - New file extending upstream image with: - libportaudio2 for audio capture - sounddevice, numpy, faster-whisper via uv - Health check for voice dependencies 2. **assets/compose/ai/compose.yml** - Updated hermes service: - Changed from image: to build: for reproducibility - Added ROCm environment variables for AMD GPU ## Deployment cd /opt/data/infra sudo nixos-rebuild switch --flake .#lazyworkhorse ## Related - Depends on: https://code.lazyworkhorse.net/gortium/compose/pulls/14
Hermes added 4 commits 2026-04-30 15:16:10 +00:00
Add AI model optimizer cron job draft and initial state files 30f8ca3863
feat(hermes): add voice support with GPU-accelerated STT 7e3afe6630 
- Dockerfile: add libportaudio2 system library for audio capture
- Dockerfile: install sounddevice, numpy, faster-whisper for local STT
- compose.yml: add ROCm environment variables for AMD GPU support

Enables local faster-whisper STT with GPU acceleration via ROCm.
feat(hermes): add Dockerfile for voice support with GPU STT 9fbfc1dc51 
- Extend upstream hermes-agent image
- Add libportaudio2 for audio capture
- Install sounddevice, numpy, faster-whisper via uv
- Add health check for voice dependencies
- Include Dockerfile.full for future complete rebuild from source
fix: keep root user in Dockerfile (match upstream image) bb64234223
Hermes closed this pull request 2026-05-09 00:22:15 +00:00

Pull request closed

Please reopen this pull request to perform a merge.
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
Hermes gortium
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: gortium/infra#25
No description provided.
Delete Branch "feat/hermes-voice-gpu-support"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?