Auphonic Integration
> Beta feature - Auphonic integration is opt-in. Enable it under Tools > Settings > General > Enable beta features. The Auphonic menu will appear after saving.
ChapterForge integrates with Auphonic for professional audio post-production. Connect your own Auphonic account to process audio using your own credits - ChapterForge does not resell or proxy Auphonic billing.
Features
- Connect your Auphonic account via OAuth 2.0 (no password stored)
- View your available credit balance before submitting jobs
- Submit singletrack and multitrack audio productions
- Six built-in presets plus your own Auphonic account presets
- Audio-only validation - video files are rejected at the source
- Automatic credit estimation with 3-minute minimum warning
- Background polling with exponential backoff until jobs complete
- Download all result files (audio, transcript, captions, stats, chapters)
- Review-before-publish workflow
- Full keyboard and screen reader accessibility
Menu Access
Auphonic menu (between View and Help):
| Menu item | Shortcut | Action |
|---|---|---|
| Connect Account | - | OAuth connect / view credit balance |
| New Production | - | Submit audio for processing |
| Job History | - | View jobs and download results |
Setup
0. Enable beta features
Open Tools > Settings, go to the General tab, and check Enable beta features (Auphonic integration). Click OK. The Auphonic menu will appear in the menu bar.
1. Get an Auphonic account
Sign up at https://auphonic.com. New accounts receive free credits.
2. Register a desktop OAuth app (admin / developer step)
Contact Auphonic support to register a desktop OAuth application. You will receive a Client ID and Client Secret. These are set per deployment:
- For development: set environment variables before launching: ``
powershell $env:AUPHONIC_CLIENT_ID = "your-client-id" $env:AUPHONIC_CLIENT_SECRET = "your-client-secret" python main.py`` - For production builds: the build pipeline injects these at build time (same pattern as the GitHub feedback token).
3. Connect from inside ChapterForge
- Open Auphonic > Connect Account
- Click Connect Auphonic - your browser opens to Auphonic's authorization page
- Log in and approve access
- Return to ChapterForge - the dialog shows your credit balance
Your OAuth token is stored encrypted in %APPDATA%\ChapterForge\auphonic_token.bin using Windows DPAPI. It is never logged or sent to any server other than Auphonic.
Submitting a Production
- Open Auphonic > New Production
- Click Browse and select an audio file
- ChapterForge validates the file with ffprobe (rejects video streams)
- Your credit estimate and available balance are shown
- Choose a preset (or your own account preset)
- Enter a title and click Submit Production
- ChapterForge submits to Auphonic and polls for completion in the background
When the job finishes you receive a notification. Open Job History to download results.
Built-in Presets
| Preset | Best for | Key settings |
|---|---|---|
| Podcast Cleanup | General podcast editing | Leveling, -16 LUFS, denoise, filtering. MP3 + stats |
| Podcast Cleanup + Transcript | Podcast with accessibility | Same + transcript HTML/TXT, SRT, WebVTT captions |
| Audiobook / ACX Draft | Audiobook submission drafts | -18 LUFS, heavy denoise, WAV + FLAC |
| Lecture Cleanup | Recorded lectures | Denoise, silence cutting, MP3 + captions |
| Meeting / Interview Multitrack | Multi-person recordings | Adaptive leveling, multitrack layout, MP3 |
| Archive Master | Long-term archival | Minimal processing, FLAC + WAV + chapters |
Audio-Only Policy
This integration is audio-only. The following are enforced at every stage:
- Only audio file extensions are accepted (MP3, WAV, FLAC, M4A, AAC, OGG, Opus, AIFF, and others)
- ffprobe inspects every file for video streams - any video stream causes rejection
- Remote URLs must use HTTPS and must not resolve to private/loopback IP ranges
- Auphonic output files containing video or audiogram formats are hidden from results even if Auphonic returns them
Credit Billing
Auphonic bills based on processed audio duration:
- Minimum charge: 3 minutes per production
- Re-running with changed settings on the same input does not trigger a new charge (per Auphonic's pricing FAQ)
- Creating a new production from the same audio does trigger a new charge
- ChapterForge shows your available balance before submission and the actual credits used after completion
Job History
Auphonic > Job History shows all submitted jobs with:
- Title, status, credits used, and submission timestamp
- Download Results - saves allowed output files to a folder you choose
- Publish - triggers publishing for jobs held at review-before-publish
Data Storage
All Auphonic job data is stored locally in %APPDATA%\ChapterForge\auphonic.db (SQLite). Nothing is sent to ChapterForge's servers.
Troubleshooting
| Problem | Solution |
|---|---|
| "Not connected to Auphonic" | Open Connect Account and click Connect |
| "Insufficient credits" | Purchase credits at auphonic.com or use a shorter file |
| "File contains a video stream" | Convert to audio-only before using (e.g., extract audio with ffmpeg) |
| Token expired | Open Connect Account - ChapterForge refreshes automatically, or reconnect |
| Job stuck in Processing | Job History > Refresh - polling continues in the background |
| OAuth credentials not configured | Set AUPHONIC_CLIENT_ID and AUPHONIC_CLIENT_SECRET environment variables |