tts_for_linux use Kokoro-FastAPI

This is an Anki addon.

Function: It allows TTS playback of local audio generated by Kokoro.

1. Kokoro-FastAPI Service: https://github.com/remsky/Kokoro-FastAPI

2. Addon Configuration:

• The URL of Kokoro-FastAPI.
• Specify the required languages and their corresponding voice.

Addendum

Unlike other operating systems that have built-in TTS engines, Linux can only achieve TTS by downloading audio files from cloud services through addons, such as:

• AwesomeTTS
• gTTS

I primarily study on Linux and frequently encounter issues with these solutions, including:

• Latency
• Inability to pronunciation after network interruption (requires restarting Anki)
• Inability to pronunciation after switching profile (requires restarting Anki)

This has often been a painful experience. Recently, after encountering Kokoro, I explored this solution. It's essentially a modification of gTTS, replacing its original audio download process from Google Translate with local service requests to Kokoro-FastAPI. This resolves the issues of high latency and network interruptions, and incidentally, also solves the profile-switching problem.

这是一个anki扩展

作用是让 tts 播放 Kokoro 生成的本地音频

1. 准备好 Kokoro-FastAPI 服务 https://github.com/remsky/Kokoro-FastAPI

2. 扩展配置

   • 填入 Kokoro-FastAPI 的 url
   • 需要的语言及其对应的voice

补充

Linux不像其他操作系统有内置的TTS引擎，所以只能通过扩展从云服务下载音频文件来实现TTS，比如：

• AwesomeTTS
• gTTS

我平时主要在Linux上学习，使用这些的方案，经常会碰到的问题是：

• 延迟
• 网络中断后再恢复不能发音（需要重启anki）
• 切换用户后不能发音（需要重启anki）

所以常常是苦不堪言，最近接触到 Kokoro 后，才探索出这个方案，其实就是基于 gTTS 修改，将其原来从 Google Translate 下载音频的部分换成从 Kokoro-FastAPI 的本地服务，解决了高延迟和网络中断的问题，顺便把切换用户的问题也解决了。