MAINT: Add audio requirements to vllm image

[gwarmstrong]

Summary by CodeRabbit

• Chores

  • Added custom Docker configuration for vllm service with audio support capabilities
  • Updated container image references to use new custom build

• Documentation

  • Updated documentation to reflect audio dependencies inclusion

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

This change introduces a custom Dockerfile that extends the official vllm/vllm-openai:v0.10.2 image with vllm[audio] dependencies. The application's container configuration is updated to use this custom image, with supporting documentation reflecting the additional dependency layer.

Changes

Cohort / File(s)	Summary
Docker Configuration dockerfiles/Dockerfile.vllm, dockerfiles/README.md	New Dockerfile extending vllm/vllm-openai:v0.10.2 with vllm[audio] dependencies; documentation updated to describe usage with additional dependencies
Container Mapping nemo_skills/__init__.py	Updated container image reference from official vllm/vllm-openai:v0.10.2 to custom dockerfile:dockerfiles/Dockerfile.vllm

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

• Verify Dockerfile syntax and base image compatibility
• Confirm container mapping reference is correctly formatted and resolves to the new Dockerfile

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'MAINT: Add audio requirements to vllm image' accurately describes the main change: adding audio dependencies (vllm[audio]) to the vllm container image through a custom Dockerfile.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch georgea/add-audio-to-vllm-requirements

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 0

🧹 Nitpick comments (2)

dockerfiles/Dockerfile.vllm (1)

1-2: Optionally avoid pip cache to keep the image smaller

The install is fine as-is, but you can reduce image size a bit by disabling the pip cache in this layer:

-FROM vllm/vllm-openai:v0.10.2
-RUN pip install "vllm[audio]"
+FROM vllm/vllm-openai:v0.10.2
+RUN pip install --no-cache-dir "vllm[audio]"

Purely cosmetic, feel free to skip if image size isn’t a concern.

dockerfiles/README.md (1)

37-39: Consider referencing Dockerfile.vllm / build flow for clarity

The description is accurate, but you might make this section more actionable by explicitly pointing to the Dockerfile and/or a build example, e.g.:

We build from dockerfiles/Dockerfile.vllm, which uses the official vllm/vllm-openai:v0.10.2 image and adds the vllm[audio] dependencies.

This keeps it aligned with the earlier examples for other images.

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between c1bf755 and e59ab4f.

📒 Files selected for processing (3)

• dockerfiles/Dockerfile.vllm (1 hunks)
• dockerfiles/README.md (1 hunks)
• nemo_skills/__init__.py (1 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (2)

• GitHub Check: pre-commit
• GitHub Check: unit-tests

🔇 Additional comments (1)

nemo_skills/__init__.py (1)

18-27: Mapping vllm to Dockerfile.vllm looks consistent

Pointing "vllm" to dockerfile:dockerfiles/Dockerfile.vllm matches how other locally-built images are wired and keeps the audio-enabled image under source control. No issues from my side.