feat: model management improvements and folder migration

[jamiepine]

Summary

• Model folder migration: users can change the models storage directory with a full-screen migration overlay showing byte-level progress. Backend handles same-filesystem renames and cross-filesystem copies with progress tracking via SSE.
• Custom models directory passed to the Python sidecar via VOICEBOX_MODELS_DIR env var, persisted in serverStore.
• Hardcoded model descriptions for all supported models, displayed above HF badges in the detail card.
• "Open Folder" button next to the storage location Change button (uses @tauri-apps/plugin-shell open).
• Removed the "Not downloaded" badge from model cards.
• Fixed server settings scroll not accounting for the audio player.
• Fixed Tauri shell open permission (".*" regex instead of true) to allow opening file paths.
• Added normalize toggle to generation settings.

Summary by CodeRabbit

Release Notes

• New Features

  • Added audio normalization option in generation settings to control output volume.
  • Introduced model storage management with ability to customize models directory and migrate existing models to a new location with progress tracking.

• Improvements

  • Refined UI layout spacing in channel list and models sections.
  • Enhanced server startup to support custom model directories.
  • Improved server status polling efficiency.

• Documentation

  • Added issue pain points analysis and updated project roadmap with new model candidates.

[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: 18bae60a-63c3-4f7e-bd73-8cadcc8bcde2

📥 Commits

Reviewing files that changed from the base of the PR and between 325714b and 3ea5877.

⛔ Files ignored due to path filters (3)

• bun.lock is excluded by !**/*.lock
• tauri/src-tauri/gen/Assets.car is excluded by !**/gen/**
• tauri/src-tauri/gen/voicebox.icns is excluded by !**/gen/**

📒 Files selected for processing (20)

• app/src/App.tsx
• app/src/components/AudioTab/AudioTab.tsx
• app/src/components/ModelsTab/ModelsTab.tsx
• app/src/components/ServerSettings/GenerationSettings.tsx
• app/src/components/ServerSettings/ModelManagement.tsx
• app/src/components/ServerTab/ServerTab.tsx
• app/src/lib/api/client.ts
• app/src/lib/api/types.ts
• app/src/lib/hooks/useGenerationForm.ts
• app/src/lib/hooks/useRestoreActiveTasks.tsx
• app/src/platform/types.ts
• app/src/stores/serverStore.ts
• backend/main.py
• backend/models.py
• docs/issue-pain-points.md
• docs/plans/PROJECT_STATUS.md
• tauri/src-tauri/src/main.rs
• tauri/src-tauri/tauri.conf.json
• tauri/src/platform/lifecycle.ts
• web/src/platform/lifecycle.ts

---

📝 Walkthrough

Walkthrough

This PR introduces audio normalization functionality across the generation pipeline, adds model cache directory management with migration capabilities including progress tracking via SSE endpoints, and updates the platform lifecycle to accept an optional modelsDir parameter throughout the frontend and backend stack.

Changes

Cohort / File(s)	Summary
Audio Normalization app/src/components/ServerSettings/GenerationSettings.tsx, app/src/lib/api/types.ts, app/src/lib/hooks/useGenerationForm.ts, backend/models.py	Added normalize boolean field to GenerationRequest, UI checkbox in GenerationSettings, and wired normalization flag through generation mutation payload.
Model Cache & Migration app/src/components/ServerSettings/ModelManagement.tsx, app/src/lib/api/client.ts, backend/main.py	Introduced model cache directory retrieval and migration endpoints (/models/cache-dir, /models/migrate, /models/migrate/progress) with SSE-based progress tracking; added UI for directory selection, migration confirmation, and progress overlay; integrated backend directory copying and size calculation utilities.
Platform Lifecycle Updates app/src/platform/types.ts, tauri/src/platform/lifecycle.ts, web/src/platform/lifecycle.ts, tauri/src-tauri/src/main.rs	Extended startServer and restartServer signatures to accept optional modelsDir parameter; propagated through Tauri/Web implementations and native Tauri handlers to set VOICEBOX_MODELS_DIR environment variable.
Server State Management app/src/stores/serverStore.ts	Added normalizeAudio and customModelsDir persisted state properties with setter functions.
Layout & UI Adjustments app/src/components/AudioTab/AudioTab.tsx, app/src/components/ModelsTab/ModelsTab.tsx, app/src/components/ServerTab/ServerTab.tsx	Removed padding from channel list container and ModelsTab root; added conditional bottom safe area padding in ServerTab based on player visibility; reformatted ternary expressions for readability.
App & API Integration app/src/App.tsx	Updated server startup to pass customModelsDir from store as second argument to startServer.
Configuration tauri/src-tauri/tauri.conf.json	Changed shell.open plugin setting from boolean true to regex pattern string ".*".
Polling Optimization app/src/lib/hooks/useRestoreActiveTasks.tsx	Increased server polling interval from 2000ms to 30000ms.
Documentation docs/issue-pain-points.md, docs/plans/PROJECT_STATUS.md	Added pain points snapshot documentation and updated project status with new model candidates (HumeAI TADA, MOSS-TTS, VoxCPM, Pocket TTS, Kokoro-82M) and expanded engine integration procedures.

Sequence Diagram(s)

sequenceDiagram
    actor User
    participant React as Client (React)
    participant API as API Client
    participant Server as Backend Server
    participant Platform as Tauri Platform

    User->>React: Click "Change" models directory
    React->>React: Show confirmation dialog
    User->>React: Confirm migration
    React->>API: migrateModels(destination)
    API->>Server: POST /models/migrate {destination}
    Server->>Server: Create progress tracker<br/>Start background migration
    Server-->>API: {source, destination}
    
    React->>Server: Subscribe to SSE /models/migrate/progress
    Server->>React: Stream progress updates<br/>(copied_bytes, filename)
    Note over React: Display progress bar<br/>& file details
    
    Server->>Server: Finish migration
    Server->>React: Final progress event
    
    React->>API: restartServer(customModelsDir)
    API->>Platform: Invoke restart_server
    Platform->>Server: Set VOICEBOX_MODELS_DIR env var<br/>Restart backend
    Server-->>Platform: Server ready
    Platform-->>API: Restart complete
    
    React->>React: Invalidate queries
    React->>User: Show success message

Loading

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~50 minutes

Possibly related PRs

• feat: chunked TTS generation for long text (engine-agnostic) #266: Modifies the same generation call sites in backend/main.py (generate, stream_speech, download_chatterbox_turbo_background) for audio normalization integration.
• feat: LuxTTS integration — multi-engine TTS support #254: Extends the model management UI in app/src/components/ServerSettings/ModelManagement.tsx with overlapping model handling features.
• feat: CUDA backend swap via binary download and restart #252: Modifies the platform lifecycle signatures (startServer/restartServer) in Tauri and related platform modules to accept additional parameters.

Poem

🐰 hops excitedly
Audio whispers, normalized and bright,
Models waltz to their new home with delight,
Directories migrate with progress in tow,
Each file finds its place, watch the bar glow! 📊✨

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch feat/model-management-improvements

📝 Coding Plan

• Generate coding plan for human review comments

---

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.}