Kokoro Web - Effortless TTS for HridaAI

What is Kokoro Web?

Kokoro Web provides a lightweight, OpenAI-compatible API for the powerful Kokoro-82M text-to-speech model, seamlessly integrating with HridaAI to enhance your AI conversations with natural-sounding voices.

🚀 Two-Step Integration

1. Deploy Kokoro Web API (One Command)

services:
  kokoro-web:
    image: ghcr.io/eduardolat/kokoro-web:latest
    ports:
      - "3000:3000"
    environment:
      # Change this to any secret key to use as your OpenAI compatible API key
      - KW_SECRET_API_KEY=your-api-key
    volumes:
      - ./kokoro-cache:/kokoro/cache
    restart: unless-stopped

Run with: docker compose up -d

2. Connect HridaAI (30 Seconds)

1. In HridaAI, go to Admin Panel → Settings → Audio
2. Configure:
   • Text-to-Speech Engine: OpenAI
   • API Base URL: http://localhost:3000/api/v1 (If using Docker: http://host.docker.internal:3000/api/v1)
   • API Key: your-api-key (from step 1)
   • TTS Model: model_q8f16 (best balance of size/quality)
   • TTS Voice: af_heart (default warm, natural english voice). You can change this to any other voice or formula from the Kokoro Web Demo

That's it! Your HridaAI now has AI voice capabilities.

🌍 Supported Languages

Kokoro Web supports 8 languages with specific voices optimized for each:

• English (US) - en-us
• English (UK) - en-gb
• Japanese - ja
• Chinese - cmn
• Spanish - es-419
• Hindi - hi
• Italian - it
• Portuguese (Brazil) - pt-br

Each language has dedicated voices for optimal pronunciation and natural flow. See the Contact Us for the complete list of language-specific voices or use the Kokoro Web Demo to preview and create your own custom voices instantly.

💾 Optimized Models for Any Hardware

Choose the model that fits your hardware needs:

Model ID	Optimization	Size	Ideal For
model_q8f16	Mixed precision	86 MB	Recommended - Best balance
model_quantized	8-bit	92.4 MB	Good CPU performance
model_uint8f16	Mixed precision	114 MB	Better quality on mid-range CPUs
model_q4f16	4-bit & fp16 weights	154 MB	Higher quality, still efficient
model_fp16	fp16	163 MB	Premium quality
model_uint8	8-bit & mixed	177 MB	Balanced option
model_q4	4-bit matmul	305 MB	High quality option
model	fp32	326 MB	Maximum quality (slower)

✨ Try Before You Install

Visit the Kokoro Web Demo to preview all voices instantly. This demo:

• Runs 100% in your browser - No server required
• Free to use - No usage limits or registration needed
• Zero installation - Just visit the website and start creating
• All features included - Test any voice or language immediately

Need More Help?

For additional options, voice customization guides, and advanced settings, visit the Contact Us.

Troubleshooting

Connection Issues

If HridaAI can't reach Kokoro Web:

• Docker Desktop (Windows/Mac): Use http://host.docker.internal:3000/api/v1
• Docker Compose (same network): Use http://kokoro-web:3000/api/v1
• Linux Docker: Use your host machine's IP address

Voice Not Working

1. Verify the secret API key matches in both the Kokoro Web config and HridaAI settings
2. Test the API directly:

   curl -X POST http://localhost:3000/api/v1/audio/speech \
     -H "Authorization: Bearer your-api-key" \
     -H "Content-Type: application/json" \
     -d '{"input": "Hello world", "voice": "af_heart"}'

For more troubleshooting tips, see the Audio Troubleshooting Guide.

Enjoy natural AI voices in your HridaAI conversations!