Text-To-Speech Serverless RunPod Worker

RunPod Endpoint

This repository contains the worker for the xTTSv2 AI Endpoints.

Docker Image

docker build .

or

docker pull devbes/tts-runpod-serverless-worker:latest

Continuous Deployment

This worker follows a modified version of the worker template where the Docker build workflow contains additional SD models to be built and pushed.

API

{
  "input": {
      "language": <language:str>,
      "voice": {
          "speaker_0": "url"
          "speaker_1": "url"
          },
      "text": [
          ["speaker_0", "text"],
          ["speaker_1", "text"],
          ...
          ["speaker_1", "text"],
      ],
      "gpt_cond_len": <gpt_cond_len:int>,
      "max_ref_len": <max_ref_len:int>,
      "speed": <speed:float>
      "enhance_audio": <enhance_audio:bool>
  }
}