In short: I wanted my blog posts to be listenable, not just readable. Built a WordPress plugin that converts posts to natural-sounding audio using ElevenLabs API. Works with multilingual sites, lets visitors play posts with one click.
The Problem: People Don’t Always Read
Let’s be honest — not everyone has time to read a 2,000-word blog post.
Some people prefer to listen while commuting, cooking, or working out. Others have visual impairments or reading difficulties. And some just find audio more engaging than text.
I looked at existing solutions:
- Manual recording — time-consuming, doesn’t scale
- Browser TTS — robotic, no personality
- Third-party services — expensive, limited control, often require embedding external players
I wanted something different: a WordPress plugin that generates natural-sounding audio for each post, in multiple languages, with full control over voices and models.
The Solution: EasyTalk + ElevenLabs API
ElevenLabs has some of the most realistic AI voices I’ve heard. Their API is straightforward, and they offer multiple models for different use cases.
So I built EasyTalk — a WordPress plugin that:
- Connects to ElevenLabs API
- Generates audio for any post or page
- Supports multilingual sites (via Translio)
- Adds a clean audio player to your content
- Shows audio availability in the posts list
Why ElevenLabs?
- Natural voices — not robotic, actual human-like speech with emotion
- Multiple models — Flash (fast & cheap), Multilingual v2 (high quality), Turbo (balanced)
- 32 languages — perfect for multilingual blogs
- Free tier — 10,000 credits/month (~10 minutes of audio) to test before committing
- Simple API — clean documentation, easy integration
What EasyTalk Does
🎙️ Voice & Model Selection
In the plugin settings, you can configure:
| Setting | Options |
|---|---|
| Voice | Choose from ElevenLabs voice library or your cloned voices |
| Model | Flash v2.5, Multilingual v2, Turbo v2.5 |
| Language | Auto-detect or manual selection |
Different voices work better for different content. A professional male voice for business articles, a friendly female voice for lifestyle content — you decide.
🌍 Full Multilingual Support
EasyTalk integrates seamlessly with Translio (my AI translation plugin for WordPress).
If your blog is in multiple languages:
- Each language version gets its own audio
- The plugin automatically detects the post language
- Generates audio with the appropriate voice/accent
Example: Your post in English gets an American voice. The Spanish translation gets a native Spanish speaker. The German version — a German voice. All automatic.
Quick visual overview of your audio content status.
▶️ Frontend Audio Player
Each post with audio gets a clean, minimal player:
- Play/pause button
- Progress bar
- Playback speed control (0.5x to 2x)
- Download option (optional)
How It Works
- Install & Configure
- Activate EasyTalk
- Enter your ElevenLabs API key
- Select default voice and model
- Generate Audio
- Open any post in the editor
- Find the “EasyTalk” metabox
- Click “Generate Audio”
- Wait 30-60 seconds (depending on post length)
- Done
- Audio player appears on the post
- Visitors can listen instead of read
Cost Breakdown
ElevenLabs Pricing
| Plan | Credits/month | Approx. Audio | Cost |
|---|---|---|---|
| Free | 10,000 | ~10 minutes | $0 |
| Starter | 30,000 | ~30 minutes | $5/month |
| Creator | 100,000 | ~100 minutes | $22/month |
| Pro | 500,000 | ~8 hours | $99/month |
What Does This Mean in Practice?
A typical 1,500-word blog post = ~10,000 characters = ~10 minutes of audio.
- Free tier: ~1 post/month for testing
- Starter ($5): ~3 posts/month
- Creator ($22): ~10 posts/month
- Pro ($99): ~50 posts/month
For most blogs publishing 2-4 posts per week, the Creator plan covers everything.
Flash Models = 50% Cheaper
Using Flash v2.5 or Turbo models costs only 0.5 credits per character — effectively doubling your quota.
Use Cases
Personal Blogs
Make your content accessible to more people. Commuters, multitaskers, and visually impaired readers will thank you.
Business & Marketing
Turn your thought leadership articles into mini-podcasts. Repurpose for social media audio clips.
Education & Documentation
Help students and learners absorb content in different ways. Audio + text = better retention.
Multilingual Publications
Serve international audiences with native-sounding voices in their language.
Technical Details
Requirements
- WordPress 5.8+
- PHP 7.4+
- ElevenLabs API key
- Translio plugin (for multilingual support)
API Integration
EasyTalk uses ElevenLabs Text-to-Speech API:
- Sends post content (stripped of HTML)
- Receives MP3 audio file
- Stores in WordPress Media Library
- Attaches to post as custom field
Performance
- Audio files are cached — generated once, served forever
- No impact on page load (audio loads asynchronously)
- CDN-compatible
Why Not Just Use Browser TTS?
Browser text-to-speech (like speechSynthesis API) is free, but:
| Browser TTS | EasyTalk + ElevenLabs | |
|---|---|---|
| Voice quality | Robotic | Natural, human-like |
| Consistency | Varies by browser/device | Same everywhere |
| Languages | Limited | 32 languages |
| Offline | No (needs browser) | Yes (audio file) |
| Customization | Minimal | Full control |
The difference is night and day. ElevenLabs voices sound like actual humans reading your content.
Summary
Problem: Not everyone reads blog posts — some prefer audio
Solution: EasyTalk WordPress plugin + ElevenLabs API
Result:
- Natural-sounding audio for any post
- Full multilingual support via Translio
- Clean audio player in your theme
- ~$22/month for most blogs (or free to test)
Want Audio on Your Website?
I’m not releasing EasyTalk publicly yet, but I can help you implement it.
If you want to add voice narration to your blog or website, reach out to me. My team and I will set up the solution for you — customized to your content, languages, and brand voice.
→ Contact me to discuss your project
What do you think about audio content on blogs? Do you prefer reading or listening? Let me know in the comments or on Telegram.
