OpenClaw Bot
9720390e1a
- Multi-provider TTS service (OpenAI, Piper, macOS say) - Supabase Storage integration for audio files - RSS 2.0 feed with iTunes extensions for podcast distribution - Web audio player at /podcast page - Integration with daily digest workflow - Manual TTS generation script - Complete documentation in PODCAST_SETUP.md
127 lines
4.4 KiB
Markdown
127 lines
4.4 KiB
Markdown
# Podcast Implementation Summary
|
|
|
|
## What Was Built
|
|
|
|
A complete podcast solution for the Daily Digest blog that automatically converts blog posts to audio using Text-to-Speech (TTS) and distributes them via RSS feed.
|
|
|
|
## Features Delivered
|
|
|
|
### 1. TTS Service (`src/lib/tts.ts`)
|
|
- **Multi-provider support**: OpenAI (paid), Piper (free), macOS say (free)
|
|
- **Text preprocessing**: Automatically strips markdown, URLs, and code blocks
|
|
- **Async generation**: Non-blocking to not delay post creation
|
|
- **Error handling**: Graceful fallback if TTS fails
|
|
|
|
### 2. Audio Storage (`src/lib/storage.ts`)
|
|
- **Supabase Storage integration**: Uses existing Supabase project
|
|
- **Automatic bucket creation**: Creates `podcast-audio` bucket if needed
|
|
- **Public access**: Audio files accessible for podcast apps
|
|
|
|
### 3. RSS Feed (`src/app/api/podcast/rss/route.ts`)
|
|
- **RSS 2.0 with iTunes extensions**: Compatible with Apple Podcasts, Spotify, Google Podcasts
|
|
- **Auto-updating**: Pulls latest episodes from database
|
|
- **Proper metadata**: Titles, descriptions, duration, publication dates
|
|
|
|
### 4. Podcast Page (`src/app/podcast/page.tsx`)
|
|
- **Web audio player**: Play episodes directly on the site
|
|
- **Episode listing**: Browse all available podcast episodes
|
|
- **Subscribe links**: RSS, Apple Podcasts, Spotify
|
|
- **Responsive design**: Works on mobile and desktop
|
|
|
|
### 5. Integration with Digest Workflow
|
|
- **Updated `/api/digest`**: Now triggers TTS generation after saving post
|
|
- **Optional audio**: Can disable with `generateAudio: false`
|
|
- **Background processing**: Doesn't block the main response
|
|
|
|
### 6. Manual TTS Script (`src/scripts/generate-tts.ts`)
|
|
- **Single post**: `npm run generate-tts -- <post_id>`
|
|
- **Batch processing**: `npm run generate-tts:all`
|
|
- **Force regeneration**: `--force` flag to overwrite existing
|
|
|
|
### 7. UI Updates
|
|
- **Blog post audio player**: Shows audio player for posts with audio
|
|
- **Podcast link in header**: Easy navigation to podcast page
|
|
- **Visual indicators**: Shows which posts have audio
|
|
|
|
## Files Created/Modified
|
|
|
|
### New Files
|
|
```
|
|
src/
|
|
├── app/
|
|
│ ├── api/podcast/rss/route.ts # RSS feed endpoint
|
|
│ └── podcast/page.tsx # Podcast web player
|
|
├── lib/
|
|
│ ├── tts.ts # TTS service abstraction
|
|
│ ├── storage.ts # Supabase storage helpers
|
|
│ └── podcast.ts # RSS generation utilities
|
|
├── scripts/
|
|
│ └── generate-tts.ts # Manual TTS generation script
|
|
└── ../PODCAST_SETUP.md # Setup documentation
|
|
```
|
|
|
|
### Modified Files
|
|
```
|
|
src/
|
|
├── app/
|
|
│ ├── api/digest/route.ts # Added TTS trigger
|
|
│ └── page.tsx # Added audio player, podcast link
|
|
├── ../package.json # Added generate-tts scripts
|
|
├── ../tsconfig.json # Added scripts to includes
|
|
└── ../.env.local # Added TTS environment variables
|
|
```
|
|
|
|
## Configuration
|
|
|
|
### Environment Variables
|
|
```bash
|
|
# Enable TTS generation
|
|
ENABLE_TTS=true
|
|
|
|
# TTS Provider: openai, piper, or macsay
|
|
TTS_PROVIDER=openai
|
|
|
|
# OpenAI settings (if using OpenAI)
|
|
OPENAI_API_KEY=sk-your-key-here
|
|
TTS_VOICE=alloy # alloy, echo, fable, onyx, nova, shimmer
|
|
|
|
# Piper settings (if using Piper)
|
|
PIPER_MODEL_PATH=./models/en_US-lessac-medium.onnx
|
|
```
|
|
|
|
### Database Schema
|
|
```sql
|
|
ALTER TABLE blog_messages
|
|
ADD COLUMN audio_url TEXT,
|
|
ADD COLUMN audio_duration INTEGER;
|
|
```
|
|
|
|
## URLs
|
|
|
|
- **RSS Feed**: `https://blog-backup-two.vercel.app/api/podcast/rss`
|
|
- **Podcast Page**: `https://blog-backup-two.vercel.app/podcast`
|
|
- **Blog**: `https://blog-backup-two.vercel.app`
|
|
|
|
## Cost Analysis
|
|
|
|
| Component | Provider | Monthly Cost |
|
|
|-----------|----------|--------------|
|
|
| TTS | OpenAI | ~$2-4 |
|
|
| TTS | Piper/macOS | $0 |
|
|
| Storage | Supabase | $0 (free tier) |
|
|
| RSS Hosting | Vercel | $0 |
|
|
| **Total** | | **$0-4/month** |
|
|
|
|
## Next Steps to Deploy
|
|
|
|
1. **Database**: Run the SQL to add `audio_url` and `audio_duration` columns
|
|
2. **Supabase Storage**: Create `podcast-audio` bucket (or let app auto-create)
|
|
3. **Environment**: Add `ENABLE_TTS=true` and `OPENAI_API_KEY` to production
|
|
4. **Deploy**: `npm run build && vercel --prod`
|
|
5. **Test**: Generate a test episode and verify RSS feed
|
|
6. **Submit**: Add RSS URL to Apple Podcasts, Spotify, etc.
|
|
|
|
## Documentation
|
|
|
|
See `PODCAST_SETUP.md` for complete setup instructions, troubleshooting, and usage guide.
|