Skip to content

Add Chinese README and Enhance Model API Response Format - #38

Open
Anyexyz wants to merge 2 commits into
travisvn:mainfrom
Anyexyz:main
Open

Add Chinese README and Enhance Model API Response Format#38
Anyexyz wants to merge 2 commits into
travisvn:mainfrom
Anyexyz:main

Conversation

@Anyexyz

@Anyexyz Anyexyz commented Aug 28, 2025

Copy link
Copy Markdown

Changes Made

This PR makes two improvements to the project:

  1. Added a comprehensive Chinese README (README.zh-CN.md) to improve accessibility for Chinese-speaking users
  2. Updated the /v1/models endpoint response format to better match the OpenAI API standard

Detailed Changes

  • Created a complete Chinese translation of the README with all key information preserved
  • Added owned_by: "openai-edge-tts" field to model data in the API response
  • Updated the get_models_formatted() function to include ownership information
  • Maintained JSON structure consistency with OpenAI standard

Benefits

  • Improves project accessibility for Chinese-speaking developers
  • Enhances compatibility with applications that rely on the OpenAI API
  • Enables seamless integration with third-party applications
  • Maintains standardized API structure and documentation

Testing

  • Verified all documentation links and formatting in the Chinese README
  • Confirmed that the updated /v1/models endpoint returns the expected format with the following structure:
{
  "object": "list",
  "data": [
    {
      "id": "tts-1",
      "object": "model",
      "owned_by": "openai-edge-tts"
    },
    {
      "id": "tts-1-hd",
      "object": "model",
      "owned_by": "openai-edge-tts"
    },
    {
      "id": "gpt-4o-mini-tts",
      "object": "model",
      "owned_by": "openai-edge-tts"
    }
  ]
}

etugrand pushed a commit to etugrand/openai-edge-tts that referenced this pull request Jun 20, 2026
…avisvn#38) into v1

- travisvn#35: fix generate_speech_stream — drive the async generator from a
  dedicated event loop instead of asyncio.run() (which raised "a coroutine
  was expected, got async_generator" and broke the SSE path, issue travisvn#34)
- travisvn#56: add correct pcm ffmpeg maps (pcm_s16le / s16le) to the file-based
  _generate_audio path; skip bitrate for pcm like wav
- travisvn#39: introduce HOST env var (default 0.0.0.0) so non-Docker installs can
  bind localhost; wired into waitress.serve()
- travisvn#38: make /models OpenAI-compatible ({"object":"list","data":[...]} with
  object/owned_by on each model)

Skipped travisvn#45 (v1 already has /health, HEALTHCHECK, curl, EXPOSE) and travisvn#59
(opt-in 60db.ai paid backend — out of scope for v1's free edge-tts focus).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant