README.md

metadata

title: Yoruba TTS API
emoji: 🗣️
colorFrom: yellow
colorTo: red
sdk: docker
app_port: 7860
pinned: false
license: cc-by-nc-4.0

Yoruba TTS API

Text-to-Speech API for Yoruba language using the facebook/mms-tts-yor model.

Model Information

• Model: facebook/mms-tts-yor
• Architecture: VITS (Variational Inference TTS)
• Parameters: 36.3M
• License: CC-BY-NC 4.0 (non-commercial use)

API Endpoints

POST /tts

Generate speech from Yoruba text.

Request:

{
  "text": "Bawo ni"
}

Response:

{
  "audio": "UklGRiQAAABXQVZFZm10...",
  "cached": false
}

The audio field contains base64-encoded WAV audio.

GET /health

Check service health.

Response:

{
  "status": "healthy",
  "model": "facebook/mms-tts-yor"
}

Usage Example

import requests
import base64

response = requests.post(
    "https://YOUR-SPACE.hf.space/tts",
    json={"text": "Bawo ni"}
)

audio_b64 = response.json()["audio"]
audio_bytes = base64.b64decode(audio_b64)

with open("output.wav", "wb") as f:
    f.write(audio_bytes)

Limitations

• Maximum text length: 500 characters
• Audio format: WAV (16-bit PCM)
• Sample rate: Model default (~22050 Hz)