setup --speakers: opt-in diarization install + actionable messages

[nmbrthirteen]

Speaker detection deps (pyannote+torch, ~2GB) are excluded from the hermetic runtime. Adds 'podcli setup --speakers' to install them on demand, and replaces dead-end 'source install only' messages with the exact command to run (setup --speakers / env set HF_TOKEN).

Summary by CodeRabbit

• New Features

  • Added --speakers flag to setup command for installing speaker diarization dependencies.

• Documentation

  • Improved error messaging with specific guidance distinguishing between missing packages versus missing authentication tokens.
  • Updated CLI help text with speaker detection setup instructions and size information.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: f0e77cdf-a97e-4c87-a83e-5e35c0207f86

📥 Commits

Reviewing files that changed from the base of the PR and between 399c3c8 and d94375f.

📒 Files selected for processing (4)

• backend/cli.py
• backend/services/speaker_detection.py
• cli/internal/provision/provision.go
• cli/main.go

---

📝 Walkthrough

Walkthrough

Adds a --speakers flag to the podcli setup CLI command backed by a new EnsureSpeakerDeps Go function that pip-installs pyannote.audio>=3.1.0 and speechbrain. Python backend error messages in run_diarization, cmd_info, and print_banner are updated to reference podcli setup --speakers and to distinguish a missing install from a missing HF_TOKEN.

Changes

Speaker Diarization Setup

Layer / File(s)	Summary
EnsureSpeakerDeps provisioner and --speakers flag cli/internal/provision/provision.go, cli/main.go	EnsureSpeakerDeps validates the hermetic Python binary, then runs pip install pyannote.audio>=3.1.0 speechbrain. The setup command gains a --speakers boolean flag that calls this function, prints a success/error message, and help text is updated with the new option.
Python backend error and status message updates backend/services/speaker_detection.py, backend/cli.py	run_diarization's ImportError now references podcli setup --speakers. cmd_info and print_banner branch on whether pyannote is installed vs. only HF_TOKEN is missing, printing targeted remediation commands for each case.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐇 Hippity-hop, a flag appears,
--speakers rings in rabbit ears!
Pip installs with a stderr note,
pyannote.audio gets its coat.
Token missing? The banner knows—
two tidy messages, and off it goes! 🎙️

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/speakers-optional-install

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 golangci-lint (2.12.2)

level=error msg="[linters_context] typechecking error: pattern ./...: directory prefix . does not contain main module or its selected dependencies"

---

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.}