60 lines
1.7 KiB
Markdown
60 lines
1.7 KiB
Markdown
# TTS Research Summary
|
|
|
|
## Recommended Approach
|
|
**Piper (Open-source) + GitHub Pages Hosting**
|
|
- Quality: High (Mimics human speech)
|
|
- Cost: Free (open-source)
|
|
- Setup: Python + pre-trained models
|
|
- Pros: No API limits, full control
|
|
- Cons: Requires technical setup
|
|
|
|
## Free TTS API Options
|
|
1. **Google Cloud TTS**
|
|
- Free tier: 1000 chars/month
|
|
- Quality: Professional-grade
|
|
- Cost: Free for small usage
|
|
- [Learn more](https://cloud.google.com/text-to-speech)
|
|
|
|
2. **AWS Polly**
|
|
- Free tier: 5 million characters/month
|
|
- Quality: Industry-leading
|
|
- Cost: Free for initial usage
|
|
- [Learn more](https://aws.amazon.com/polly/)
|
|
|
|
3. **ElevenLabs**
|
|
- Free tier: 1000 credits/month
|
|
- Quality: High (emotional voices)
|
|
- Cost: Free for basic use
|
|
- [Learn more](https://elevenlabs.io/)
|
|
|
|
## Browser-Based TTS
|
|
- Chrome's built-in TTS
|
|
- [Balabolka](https://www.balabolka.net/) (Windows)
|
|
- [Lingua](https://lingua.io/) (browser extension)
|
|
|
|
## RSS Feed Generation
|
|
- [Podbean](https://podbean.com/)
|
|
- [RSS2Email](https://www.rss2email.com/)
|
|
- [FeedBurner](https://feedburner.google.com/)
|
|
|
|
## Audio Hosting Options
|
|
1. **GitHub Pages** (free, static files)
|
|
2. **Dropbox** (free 2GB storage)
|
|
3. **SoundCloud** (free, streaming)
|
|
4. **Amazon S3** (free tier: 5GB storage)
|
|
|
|
## Sample Implementation
|
|
```bash
|
|
# Install Piper
|
|
pip install piper
|
|
|
|
# Generate audio
|
|
piper --model en_US-ljspeech --text "Your blog post text here" --output audio.mp3
|
|
|
|
# Host on GitHub Pages
|
|
# 1. Create GitHub repo
|
|
# 2. Upload audio.mp3
|
|
# 3. Enable GitHub Pages in settings
|
|
```
|
|
|
|
**Note**: Web search results were not available due to missing Brave API key. This document uses general knowledge. Consider configuring the API key for more detailed research. |