Apppublic

Ichika065/Esp32Ai-docker-test

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes

App README

🎤 Kokoro TTS API

A fast Text-to-Speech API using Kokoro running on Hugging Face Docker Spaces.

🚀 Live API

After deployment, your API will be available at:

—Base URL: /
—Docs: /docs
—Health: /health
—TTS Endpoint: /v1/audio/speech

🌐 Example Space URL

https://YOUR_SPACE.hf.space

📡 TTS API Usage

POST `/v1/audio/speech`

Request:

json

{
  "input": "Hello world",
  "voice": "af_heart"
}

---

## ⚡ CPU Speed Tips (Hugging Face Free)

- Request text is normalized and trimmed for faster generation.
- Responses are cached in-memory for repeated `(input, voice)` calls.
- WAV encoding uses lightweight stdlib encoding for lower overhead.
- Runtime threads are limited to avoid CPU oversubscription.

Optional environment variable:

- `MAX_CHARS=350` (lower this to make responses faster on tiny CPU)