Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/rerender2021/echo
A simple asr translator powered by avernakis react.
https://github.com/rerender2021/echo
asr ave avernakis nlp offline translation
Last synced: about 1 month ago
JSON representation
A simple asr translator powered by avernakis react.
- Host: GitHub
- URL: https://github.com/rerender2021/echo
- Owner: rerender2021
- License: mit
- Created: 2023-03-11T13:26:21.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2023-11-12T10:43:08.000Z (about 1 year ago)
- Last Synced: 2024-08-02T16:39:37.575Z (4 months ago)
- Topics: asr, ave, avernakis, nlp, offline, translation
- Language: TypeScript
- Homepage: https://rerender2021.github.io/products/echo/
- Size: 13.8 MB
- Stars: 101
- Watchers: 1
- Forks: 9
- Open Issues: 9
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- gitcat - echo
README
[![build](https://github.com/rerender2021/echo/actions/workflows/build.yml/badge.svg?branch=main&event=push)](https://github.com/rerender2021/echo/actions/workflows/build.yml) [![pack](https://github.com/rerender2021/echo/actions/workflows/pack.yml/badge.svg?branch=main&event=push)](https://github.com/rerender2021/echo/actions/workflows/pack.yml)
# 简介回声 (Echo) 是一个简单的翻译器,原理:
- 使用语音识别,获得文字用于翻译。目前支持离线情况下,英文翻译成中文。
- GUI 部分则是使用 [Ave React](https://qber-soft.github.io/Ave-React-Docs/) 开发的。![echo-usage](./docs/images/echo-usage.png)
演示视频见:
- v1.0.0: [回声:实时英语语音翻译](https://www.bilibili.com/video/BV11L411d7HE/)
- v1.1.0: [回声更新:支持使用GPU & 长句分解](https://www.bilibili.com/video/BV1Qa4y1M7jV/)
- v1.2.0: [回声更新:支持历史字幕 & 自助问题排查](https://www.bilibili.com/video/BV1XN411g7tF/)
# 使用说明
- 软件首页:https://rerender2021.github.io/products/echo/
# 开发者向
## 本地开发
```bash
> npm install
> npm run dev
```开发过程中需要确保本机启动了语音识别服务器和翻译服务器。
- 语音识别服务器:[ASR-API 1.1.0](https://github.com/rerender2021/ASR-API/releases/download/1.1.0/asr-server-v1.1.0.zip)
- 翻译服务器:[NLP-API 1.0.1](https://github.com/rerender2021/NLP-API/releases/download/1.0.1/NLP-API-v1.0.1.zip)下载它们,并解压到项目下,确保项目目录结构如下:
```
- nlp-server
- NLP-API.exe
- ...
- asr-server-v1.1.0
- ASR-API.exe
- ...
- src
- ...
- package.json
```如需使用GPU:
- GPU翻译服务器:下载链接中的2个压缩分卷并解压缩(文件太大,只能分卷压缩上传)
- [NLP-GPU-API 1.0.0](https://github.com/rerender2021/NLP-GPU-API/releases/tag/1.0.0)下载后,解压到项目下,确保项目目录结构如下:
```
- nlp-gpu-server
- NLP-GPU-API.exe
- ...
- asr-server-v1.1.0
- ASR-API.exe
- ...
- src
- ...
- package.json
```## 功能扩展
运行过程中,语音识别和翻译会请求本地接口,因此,不使用以上离线服务器,而是自己起一个服务器对接在线 API,也可正常使用。
相关接口和数据结构约定见代码:
- 语音识别: [./src/asr/asr.ts](./src/asr/asr.ts)
- 翻译: [./src/nlp/helsinki-nlp.ts](./src/nlp/helsinki-nlp.ts)## 打包发布
- 生成 exe
```bash
> npm run release
```# 开源协议
[MIT](./LICENSE)
# 赞赏
`:)` 如果此软件值得赞赏,可以请作者看小说,一元足足可看八章呢。