AI Agent Hub
Back to skills
Doubao TTS Speech Synthesis icon

Doubao TTS Speech Synthesis

Design & Media Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_f95f21b8/dlazy-doubao-tts according to https://skillhub.cn/install/skillhub.md.

About this skill

Problem

When a workflow needs text broadcast, speech generation, or TTS integration, calling a cloud API directly means handling credentials, requests, polling, and error codes. This skill wraps doubao tts as an Agent-oriented command entry point, making text-to-speech a step in an automation pipeline.

How It Works

It is built on the dLazy hosted API and centers on running dlazy doubao-tts. Key steps:
- Configure the API key through the CLI, or provide it via the DLAZY_API_KEY environment variable.
- Pass prompt, voice, and related parameters; local file paths are uploaded by the CLI to dLazy media storage for model access.
- Synchronous calls return generated output URLs; with --no-wait, the response includes generateId and status, so poll the status endpoint.
- Output includes ok, error codes, and messages such as unauthorized or insufficient_balance, which helps agents branch on failure.

Boundaries

The skill is described as a thin client; it does not perform extra local resource access beyond the dLazy CLI. Requests, files, and generated outputs go through dLazy endpoints. Insufficient balance returns a recharge prompt, and async failures may include backend reasons such as safety policy violations. It is useful for debugging and integrating speech generation, not for local offline TTS.

Use Cases

  • In an automated broadcast pipeline, synthesize a news digest into a speech URL for review and publishing.
  • While debugging an agent voice task, verify API key handling, insufficient balance, and async `generateId` polling.
  • In a content pipeline, pass upstream output parameters to `dlazy doubao-tts` to generate narration audio.
  • When creating voice script samples, generate natural speech quickly and retrieve the `files.dlazy.com` audio link.

Best For

  • Backend engineers building voice automation who need to call Doubao TTS from agent workflows and handle error codes.
  • Agent application developers maintaining dLazy CLI workflows who need to insert a text-to-speech step into a command pipeline.
  • QA engineers testing TTS APIs who need to verify unauthorized, insufficient balance, and async task failure paths.
  • Content engineers producing voice samples who need to turn script text into referenceable audio URLs quickly.