kesha-voice-kit

分类: 开发与编程 | 上传者: drakulavich | 下载: 0 | 版本: v1.0（最新）

语音消息的本地语音到文本。完全在您的计算机上运行 - 没有 API 密钥，没有云。

更新日志: Source: GitHub https://github.com/drakulavich/kesha-voice-kit

目录结构

当前层级: tree/main/

📁 .claude/
- 📄 settings.json 833 B
📁 .github/
- 📁 actions/
  - 📁 install-parakeet-backend/
    
    📄 action.yml 1.4 KB
  - 📁 setup-bun/
    
    📄 action.yml 656 B
- 📁 scripts/
  - 📄 generate-pr-comment.ts 1.3 KB
  - 📄 junit-to-markdown.ts 4.1 KB
- 📁 workflows/
  - 📄 build-engine.yml 1.9 KB
  - 📄 ci.yml 3.9 KB
  - 📄 claude-code-review.yml 1.4 KB
  - 📄 claude.yml 1.8 KB
  - 📄 rust-test.yml 652 B
- 📄 dependabot.yml 386 B
📁 assets/
- 📄 benchmark.svg 129 B
- 📄 logo.png 131 B
📁 bin/
- 📄 kesha.js 76 B
📁 docs/
- 📁 superpowers/
  - 📁 plans/
    
    📄 2026-04-06-lib-split.md 7.2 KB
    
    📄 2026-04-07-coreml-backend.md 19.2 KB
    
    📄 2026-04-09-cli-help-citty.md 10.7 KB
    
    📄 2026-04-09-dx-improvements.md 20.6 KB
    
    📄 2026-04-09-ffmpeg-error-dx.md 5.1 KB
    
    📄 2026-04-09-lang-detection.md 8.9 KB
    
    📄 2026-04-12-coreml-lang-detection.md 34.0 KB
    
    📄 2026-04-14-benchmark.md 15.3 KB
    
    📄 2026-04-14-rust-engine.md 32.2 KB
    
    📄 2026-04-14-ts-migration.md 13.0 KB
  - 📁 specs/
    
    📄 2026-04-06-lib-split-design.md 3.8 KB
    
    📄 2026-04-07-coreml-backend-design.md 5.9 KB
    
    📄 2026-04-07-scripts-ts-rewrite-design.md 2.0 KB
    
    📄 2026-04-09-cli-help-design.md 3.9 KB
    
    📄 2026-04-09-dx-improvements-design.md 6.8 KB
    
    📄 2026-04-09-ffmpeg-error-dx-design.md 1.7 KB
    
    📄 2026-04-09-lang-detection-design.md 2.3 KB
    
    📄 2026-04-12-coreml-lang-detection-design.md 6.5 KB
    
    📄 2026-04-14-benchmark-design.md 4.1 KB
    
    📄 2026-04-14-rust-engine-design.md 7.9 KB
📁 fixtures/
- 📁 benchmark/
  - 📄 01-ne-nuzhno-slat-soobshcheniya.ogg 130 B
  - 📄 02-prover-vse-svoi-konfigi.ogg 130 B
  - 📄 03-ty-dobavil-sebe-v-pamyat.ogg 130 B
  - 📄 04-pokazhi-ego-yuzerneim.ogg 130 B
  - 📄 05-vynesi-eshche-sekret-ot-kloda.ogg 130 B
  - 📄 06-kakie-eshche-telegram-yuzery.ogg 130 B
  - 📄 07-to-chto-nakhoditsya-v-papke.ogg 130 B
  - 📄 08-uznai-vtorogo-yuzera.ogg 130 B
  - 📄 09-ustanovi-poka-klod-kod.ogg 130 B
  - 📄 10-zakomit-izmeneniya-v-git.ogg 129 B
- 📁 benchmark-en/
  - 📄 01-check-email.ogg 38.9 KB
  - 📄 02-meeting-rescheduled.ogg 41.4 KB
  - 📄 03-review-pull-request.ogg 37.8 KB
  - 📄 04-deploy-staging.ogg 42.1 KB
  - 📄 05-database-migration.ogg 44.9 KB
  - 📄 06-code-review-session.ogg 40.7 KB
  - 📄 07-run-test-suite.ogg 39.1 KB
  - 📄 08-update-documentation.ogg 44.4 KB
  - 📄 09-refactor-notifications.ogg 44.2 KB
  - 📄 10-load-balancer-config.ogg 47.3 KB
- 📄 hello-english.oga 130 B
- 📄 hello-english.wav 130 B
- 📄 hello-spanish.wav 130 B
- 📄 silence.wav 130 B
- 📄 test-vocab.txt 52 B
📁 rust/
- 📁 src/
  - 📁 backend/
    
    📄 fluidaudio.rs 522 B
    
    📄 mod.rs 736 B
    
    📄 onnx.rs 15.4 KB
  - 📄 audio.rs 8.1 KB
  - 📄 capabilities.rs 651 B
  - 📄 lang_id.rs 2.0 KB
  - 📄 main.rs 2.0 KB
  - 📄 models.rs 3.0 KB
  - 📄 text_lang.rs 1.4 KB
  - 📄 transcribe.rs 498 B
- 📄 Cargo.lock 42.4 KB
- 📄 Cargo.toml 862 B
📁 scripts/
- 📄 benchmark.ts 11.8 KB
- 📄 convert-lang-id-model.py 8.0 KB
- 📄 postinstall.cjs 320 B
- 📄 smoke-test.ts 2.7 KB
📁 src/
- 📄 cli.ts 6.9 KB
- 📄 engine-install.ts 2.4 KB
- 📄 engine.ts 2.5 KB
- 📄 lang-id.ts 127 B
- 📄 lib.ts 527 B
- 📄 log.ts 319 B
- 📄 models.ts 176 B
- 📄 progress.ts 2.2 KB
- 📄 status.ts 1.5 KB
- 📄 suggest-command.ts 736 B
- 📄 transcribe.ts 1006 B
📁 tests/
- 📁 integration/
  - 📄 e2e-cli.test.ts 2.6 KB
  - 📄 e2e-engine.test.ts 5.3 KB
- 📁 unit/
  - 📄 cli.test.ts 6.5 KB
  - 📄 engine.test.ts 897 B
  - 📄 lib.test.ts 265 B
  - 📄 progress.test.ts 3.1 KB
  - 📄 status.test.ts 829 B
  - 📄 suggest-command.test.ts 1.3 KB
📄 .gitattributes 444 B
📄 .gitignore 138 B
📄 .npmignore 63 B
📄 AGENTS.md 4.2 KB
📄 BENCHMARK.md 6.1 KB
📄 bun.lock 2.9 KB
📄 CLAUDE.md 8.4 KB
📄 CONTRIBUTING.md 2.8 KB
📄 LICENSE 1.0 KB
📄 Makefile 880 B
📄 openclaw.plugin.json 527 B
📄 package.json 1.5 KB
📄 README.md 5.4 KB
📄 SECURITY.md 619 B
📄 SKILL.md 1.5 KB
📄 tsconfig.json 278 B

SKILL.md

# Kesha Voice Kit

Local speech-to-text for voice messages. Runs entirely on your machine — no API keys, no cloud.

## What it does

When a voice message arrives (Telegram, WhatsApp, Slack), Kesha transcribes it and returns JSON with the transcript and detected language:

```json
[{
  "file": "voice.ogg",
  "text": "Привет, как дела?",
  "lang": "ru",
  "textLanguage": { "code": "ru", "confidence": 0.99 }
}]
```

Use the `lang` field to detect which language the user spoke and respond accordingly.

## Setup

After installing the plugin, run once to download the engine and models:

```bash
kesha install
```

## Plain text mode

If you don't need language detection, switch to plain text output in your config:

```json
{
  "type": "cli",
  "command": "kesha",
  "args": ["{{MediaPath}}"]
}
```

This returns just the transcript text without JSON wrapping.

## Available flags

- `--json` — JSON output with language detection (default in this plugin)
- `--verbose` — show language detection details (audio + text language with confidence)
- `--lang <code>` — warn if detected language differs from expected (e.g. `--lang en`)

## Supported languages

Speech-to-text: Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, Ukrainian.

## Performance

~19x faster than Whisper on Apple Silicon (CoreML), ~2.5x faster on CPU (ONNX).

登录后下载/点赞/收藏 ❤ 6 | ★ 0

kesha-voice-kit

目录结构

SKILL.md

举报内容

提示