gortium/compose
1
1
Fork 0
Code Issues Pull Requests 18 Actions Packages Projects Releases Wiki Activity

feat(hermes): Piper TTS (local US male, no cloud) #17

Merged
gortium merged 25 commits from feat/voice-support-v2 into master 2026-05-09 19:39:12 +00:00
Conversation 2 Commits 25 Files Changed 4 +282 -55
Hermes commented 2026-05-09 00:21:18 +00:00
Collaborator
Copy Link

Use official Hermes Agent image as base (nousresearch/hermes-agent:latest). Add Piper TTS, remove Edge TTS. See commit messages for details.

Use official Hermes Agent image as base (nousresearch/hermes-agent:latest). Add Piper TTS, remove Edge TTS. See commit messages for details.
Hermes added 1 commit 2026-05-09 00:21:19 +00:00
feat: add ROCm GPU env vars to hermes service for faster-whisper STT 5c504501d3
Hermes added 1 commit 2026-05-09 02:38:41 +00:00
fix: replace Dockerfile with simplified stable-slim version f5171a7d6e
Hermes changed title from feat(hermes): add ROCm GPU env vars for faster-whisper STT to feat(hermes): simplified Dockerfile + ROCm GPU env vars for STT 2026-05-09 02:38:56 +00:00
Hermes added 1 commit 2026-05-09 04:09:59 +00:00
feat: add ROCm + Coqui TTS with GPU support to Dockerfile c2471818b2
Hermes changed title from feat(hermes): simplified Dockerfile + ROCm GPU env vars for STT to feat(hermes): ROCm GPU + Coqui TTS + faster-whisper STT 2026-05-09 04:13:33 +00:00
Hermes added 1 commit 2026-05-09 13:24:11 +00:00
fix: replace Coqui/ROCm with Piper TTS (simpler, local, CPU) 28213eec5c
Hermes changed title from feat(hermes): ROCm GPU + Coqui TTS + faster-whisper STT to feat(hermes): Piper TTS (local, US male) - replaces Coqui/ROCm/Edge 2026-05-09 13:24:28 +00:00
Hermes changed title from feat(hermes): Piper TTS (local, US male) - replaces Coqui/ROCm/Edge to feat(hermes): Piper TTS (local US male, no cloud) 2026-05-09 13:34:56 +00:00
Hermes added 1 commit 2026-05-09 13:41:40 +00:00
fix: clean Dockerfile with Piper TTS, external patch script 25d7611043
Hermes added 1 commit 2026-05-09 13:41:55 +00:00
chore: remove pycache e779818e73
gortium reviewed 2026-05-09 13:45:41 +00:00
ai/compose.yml Outdated
@@ -38,7 +38,13 @@ services:
- API_SERVER_HOST=0.0.0.0
- API_SERVER_KEY=hermes_local_key
- GATEWAY_ALLOW_ALL_USERS=true
- OPENROUTER_API_KEY=${OPENROUTER_API_KEY}
gortium commented 2026-05-09 13:45:41 +00:00
Owner
Copy Link

This is the right line

This is the right line
gortium marked this conversation as resolved
gortium reviewed 2026-05-09 13:46:01 +00:00
ai/compose.yml Outdated
@@ -39,3 +39,3 @@
- API_SERVER_KEY=hermes_local_key
- GATEWAY_ALLOW_ALL_USERS=true
- OPENROUTER_API_KEY=${OPENROUTER_API_KEY}
- OPENROUTER_API_KEY=${OPEN...KEY}
gortium commented 2026-05-09 13:46:01 +00:00
Owner
Copy Link

Delete this one

Delete this one
Hermes added 1 commit 2026-05-09 13:55:40 +00:00
fix: use full OPENROUTER_API_KEY variable name 78f499bde8
Hermes added 1 commit 2026-05-09 13:59:12 +00:00
fix: clean patch script - only target Edge, no Coqui references 3f080da35e
Hermes added 1 commit 2026-05-09 14:09:53 +00:00
fix: Dockerfile heredoc for voice download instead of multi-line -c 77fe8133ae
Hermes added 1 commit 2026-05-09 14:12:09 +00:00
fix: correct COPY path for patch_tts_tool.py (build context is ai/) b3fa424661
Hermes added 1 commit 2026-05-09 14:14:56 +00:00
fix: add ca-certificates for HuggingFace download 0a9507de65
Hermes added 1 commit 2026-05-09 14:20:48 +00:00
feat: switch to Norman voice (US male, medium) b750d26d80
Hermes added 1 commit 2026-05-09 14:27:11 +00:00
fix: patch source tts_tool.py path, not site-packages 3016d0da2c
Hermes added 1 commit 2026-05-09 14:28:37 +00:00
fix: remove patch step from Dockerfile (build context is just ai/) 8e9a75fe5c
Hermes added 1 commit 2026-05-09 15:21:52 +00:00
feat: switch back to Ryan high quality voice 90e227bc4e
Hermes added 1 commit 2026-05-09 15:50:38 +00:00
fix: add atomic write permission fix (preserves file mode on os.replace) 1a1cfec80a
Hermes added 1 commit 2026-05-09 16:04:34 +00:00
fix: add startup permission fix for data volume (chown critical dirs on boot) d97f1cb1e5
Hermes added 1 commit 2026-05-09 17:13:10 +00:00
fix: reinstate tts_tool.py patch step in Dockerfile 0609720b33 
Commit 8e9a75f removed the COPY+RUN of patch_tts_tool.py
because the build context was thought to be insufficient.
The build context is ai/ which contains both the Dockerfile
and patch_tts_tool.py, so COPY works fine.

Without this step the tts_tool.py silently falls through
to Edge TTS as its default provider even when
config.yaml says provider: piper, because 'piper' is not
a recognized provider in the unpatched code. This caused
the female Edge TTS voice (AriaNeural) instead of the
configured Ryan High male voice.
Hermes added 2 commits 2026-05-09 17:37:38 +00:00
fix: move TTS patch from build-time to runtime cfa2a898c3 
The build-time COPY+RUN of patch_tts_tool.py failed because
the Dockerfile starts from debian:stable-slim and only copies
the ai/ build context — there's no tools/tts_tool.py in the
image at build time (Hermes is on the mounted data volume).

Move patching to fix-permissions.sh which runs at container
startup when the data volume is mounted, so tts_tool.py is
available via the venv site-packages.

Also make patch_tts_tool.py robust: searches multiple paths
for tts_tool.py, accepts path as argument, exits 0 instead
of 1 when file/pattern not found (build must not fail).
fix: install hermes-agent from pip so build-time TTS patch works a40e347dfa 
The Dockerfile starts from debian:stable-slim, not from the official
Hermes image. Without installing hermes-agent from pip, there is no
tools/tts_tool.py in the image at build time, so the patch script
crashes with FileNotFoundError.

Adding hermes-agent to uv pip install gives us tts_tool.py in the
venv site-packages, so the COPY+RUN patch step works cleanly.

Also keep the runtime fallback in fix-permissions.sh for cases where
the volume's site-packages differ from the image.
Hermes added 1 commit 2026-05-09 17:39:31 +00:00
refactor: use official Hermes Agent image as base, not debian:stable-slim 98216d2872 
Starting from debian:stable-slim required re-installing everything
(Hermes source, Node.js, Playwright, etc.) which was redundant
and fragile. The official nousresearch/hermes-agent image already
has all that.

Now the Dockerfile:
- FROM nousresearch/hermes-agent:latest (has tts_tool.py, Playwright, etc.)
- Install Piper + voice model on top
- Patch tts_tool.py at build time (Edge fallback -> Piper)
- Runtime fallback in fix-permissions.sh for volume resilience

Cleaner, smaller Dockerfile, and the build-time patch can find
tts_tool.py because it's in the base image's venv.
Hermes added 1 commit 2026-05-09 17:45:02 +00:00
fix: install into existing venv instead of recreating it 6f17743667 
The nousresearch/hermes-agent:latest base image already has a
venv with hermes-agent installed at /opt/hermes/.venv/.
Running 'uv venv' on top of it either fails or wipes the
existing install.

Fix: activate the existing venv first, then pip install into it.
Hermes added 1 commit 2026-05-09 17:47:45 +00:00
fix: install piper-tts as root (venv is root-owned in base image) 3f80744ebd 
The nousresearch/hermes-agent:latest image creates its venv
as root. Running 'uv pip install' as USER hermes fails with
Permission denied on the site-packages directory.

Fix: keep USER root while modifying the venv, then switch
back to USER hermes for runtime.
Hermes added 1 commit 2026-05-09 19:03:13 +00:00
fix: update TTS patch for latest hermes-agent tts_tool.py 748b5037b9 
- Patch now matches the current tts_tool.py (newer version ships in
  nousresearch/hermes-agent:latest with different Edge fallback text)
- Adds dedicated elif provider == 'piper' block before else:
- Replaces else: fallback to use Piper instead of Edge
- Patches ALL copies (venv site-packages + /opt/hermes/tools/)
- Removes Edge TTS entirely as default/provider
Hermes added 1 commit 2026-05-09 19:18:20 +00:00
chore: restore system packages lost in base image migration b89be7b8f4 
The migration from debian:stable-slim to nousresearch/hermes-agent:latest
dropped several packages that were previously installed. This restores:

- poppler-utils, imagemagick (PDF/image processing)
- texlive-latex-base, latex-extra, fonts-recommended, xetex, science
- qemu-user-static, binfmt-support (cross-compilation)
- emacs-nox (text editing)

These were added in PRs 3/5, 4/5, 5/5 and earlier commits of the
compose repo. The official image already has git, curl, ffmpeg,
python3, gcc, openssh, ripgrep, tini, docker-cli, etc.
gortium merged commit 6e540635bf into master 2026-05-09 19:39:12 +00:00
gortium deleted branch feat/voice-support-v2 2026-05-09 19:39:21 +00:00
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
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: gortium/compose#17
No description provided.
Delete Branch "feat/voice-support-v2"

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?