@drakulavich/kesha-voice-kit 1.4.4 → 1.5.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/NOTICES.md +4 -10
  2. package/SKILL.md +10 -11
  3. package/package.json +2 -2
package/NOTICES.md CHANGED
@@ -27,19 +27,13 @@ Pinned to the `v6.2.1` tag.
27
27
 
28
28
  Text-to-speech (English). Apache 2.0. ONNX export by [onnx-community/Kokoro-82M-v1.0-ONNX](https://huggingface.co/onnx-community/Kokoro-82M-v1.0-ONNX).
29
29
 
30
- ### Piper voices (Russian: `ru-denis`)
30
+ ### Vosk-TTS (Russian, multi-speaker)
31
31
 
32
- Text-to-speech (multilingual, Piper VITS). MIT. Voice packs from [rhasspy/piper-voices](https://huggingface.co/rhasspy/piper-voices).
32
+ Text-to-speech (Russian, 5 baked-in speakers). Apache 2.0. Model: [alphacep/vosk-tts](https://huggingface.co/alphacep/vosk-tts) (`vosk-model-tts-ru-0.9-multi`). Integrated via [`vosk-tts-rs`](https://github.com/alphacep/vosk-tts-rs).
33
33
 
34
- ### CharsiuG2P ByT5-tiny (ONNX export)
34
+ ### misaki-rs (English G2P)
35
35
 
36
- Grapheme-to-phoneme conversion for Kokoro and Piper pipelines (#123).
37
-
38
- **License: CC-BY 4.0** (declared on the ONNX-export repo). Attribution required.
39
-
40
- - ONNX export: [klebster/g2p_multilingual_byT5_tiny_onnx](https://huggingface.co/klebster/g2p_multilingual_byT5_tiny_onnx) by Kleber Noel — CC-BY 4.0.
41
- - Base checkpoint: [charsiu/g2p_multilingual_byT5_tiny_16_layers_100](https://huggingface.co/charsiu/g2p_multilingual_byT5_tiny_16_layers_100) — **no explicit license declared on HuggingFace**; usage here follows the CC-BY 4.0 declared by the downstream export and the paper's terms.
42
- - Paper: Zhu, J., Zhang, C., & Jurgens, D. (2022). "ByT5 model for massively multilingual grapheme-to-phoneme conversion." [arXiv:2204.03067](https://arxiv.org/abs/2204.03067) · Interspeech 2022.
36
+ Embedded grapheme-to-phoneme for English (Kokoro pipeline, #207). MIT. Source: [`misaki-rs`](https://github.com/MicheleYin/misaki-rs).
43
37
 
44
38
  ## System libraries linked by the engine binary
45
39
 
package/SKILL.md CHANGED
@@ -1,15 +1,14 @@
1
1
  ---
2
2
  name: kesha-voice-kit
3
- description: Local multilingual voice toolkit — speech-to-text (STT), text-to-speech (TTS), and language detection. Runs entirely offline on Apple Silicon, Linux, and Windows. No API keys, no cloud. NVIDIA Parakeet TDT for STT across 25 European languages, Kokoro-82M + Piper VITS for TTS, plus macOS AVSpeechSynthesizer for ~180 system voices with zero install.
3
+ description: Local multilingual voice toolkit — speech-to-text (STT), text-to-speech (TTS), and language detection. Runs entirely offline on Apple Silicon, Linux, and Windows. No API keys, no cloud. NVIDIA Parakeet TDT for STT across 25 European languages, Kokoro-82M + Vosk-TTS for TTS, plus macOS AVSpeechSynthesizer for ~180 system voices with zero install.
4
4
  emoji: 🎙️
5
5
 
6
6
  requires:
7
7
  bins: [kesha]
8
8
 
9
9
  install:
10
- - kind: npm
11
- packages: [@drakulavich/kesha-voice-kit]
12
- flags: [--global]
10
+ - kind: bash
11
+ cmd: bun add -g "@drakulavich/kesha-voice-kit"
13
12
  - kind: bash
14
13
  cmd: kesha install
15
14
  ---
@@ -23,7 +22,7 @@ Local voice toolkit: transcribe voice messages to text, synthesize speech, detec
23
22
  ## When to use
24
23
 
25
24
  - **Voice memo arrived** (Telegram, WhatsApp, Slack, Signal .ogg/.opus/.m4a): transcribe with `kesha --json <path>` and branch on the detected language.
26
- - **Need to reply with audio**: synthesize with `kesha say "<text>" > reply.wav`. Auto-routes by detected language (Kokoro-82M for English, Piper for Russian). For other languages and ~180 more voices use `--voice macos-*` on macOS (zero model download).
25
+ - **Need to reply with audio**: synthesize with `kesha say "<text>" > reply.wav`. Auto-routes by detected language (Kokoro-82M for English, Vosk-TTS for Russian). For other languages and ~180 more voices use `--voice macos-*` on macOS (zero model download).
27
26
  - **Need to detect what language a file is in** before choosing a pipeline: `kesha --json audio.ogg` returns both audio-based and text-based language detection with confidence scores.
28
27
 
29
28
  ## STT: transcribe audio
@@ -57,9 +56,9 @@ Use `lang` (or the more detailed `audioLanguage`/`textLanguage`) to decide how t
57
56
 
58
57
  ```bash
59
58
  kesha say "Hello, world" > hello.wav # auto-routes en → Kokoro-82M
60
- kesha say "Привет, мир" > privet.wav # auto-routes ru → Piper
59
+ kesha say "Привет, мир" > privet.wav # auto-routes ru → Vosk-TTS
61
60
  kesha say --voice macos-de-DE "Guten Tag" > de.wav # any macOS system voice — German, French, Italian, ...
62
- kesha say --list-voices # Kokoro + Piper + ~180 macos-* voices
61
+ kesha say --list-voices # Kokoro + Vosk-TTS + ~180 macos-* voices
63
62
  ```
64
63
 
65
64
  Output: WAV mono float32. `--out <path>` writes to a file instead of stdout.
@@ -71,18 +70,18 @@ Output: WAV mono float32. `--out <path>` writes to a file instead of stdout.
71
70
  ## Install
72
71
 
73
72
  ```bash
74
- bun add --global @drakulavich/kesha-voice-kit # or: npm i -g @drakulavich/kesha-voice-kit
73
+ bun add -g @drakulavich/kesha-voice-kit # global CLI install
75
74
  kesha install # downloads engine (~350 MB)
76
- kesha install --tts # adds Kokoro + Piper RU + ONNX G2P (~490 MB more, for TTS)
75
+ kesha install --tts # adds Kokoro + Vosk-TTS RU (~990 MB more, for TTS)
77
76
  ```
78
77
 
79
- No system deps — G2P runs as ONNX alongside Kokoro/Piper. `macos-*` voices need no install either — they use voices already on the Mac.
78
+ No system deps — English G2P is embedded (`misaki-rs`); Russian G2P is bundled inside Vosk-TTS. `macos-*` voices need no install either — they use voices already on the Mac.
80
79
 
81
80
  ## Supported languages
82
81
 
83
82
  **Speech-to-text (25):** 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.
84
83
 
85
- **Text-to-speech:** English (Kokoro-82M, ~70 voices), Russian (Piper `ru-denis`), plus any macOS system voice via `--voice macos-*`.
84
+ **Text-to-speech:** English (Kokoro-82M, ~70 voices), Russian (Vosk-TTS, 5 baked-in speakers — default `ru-vosk-m02`), plus any macOS system voice via `--voice macos-*`.
86
85
 
87
86
  ## Performance
88
87
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@drakulavich/kesha-voice-kit",
3
- "version": "1.4.4",
3
+ "version": "1.5.0",
4
4
  "description": "Open-source voice toolkit. Speech-to-text (25 langs, ~19x faster than Whisper on Apple Silicon, ONNX fallback on CPU), text-to-speech (Kokoro + Vosk-TTS + ~180 macOS system voices, SSML), voice-activity detection, language detection (107 langs). Rust engine, OpenClaw skill.",
5
5
  "type": "module",
6
6
  "bin": {
@@ -30,7 +30,7 @@
30
30
  ]
31
31
  },
32
32
  "keshaEngine": {
33
- "version": "1.4.1"
33
+ "version": "1.5.0"
34
34
  },
35
35
  "scripts": {
36
36
  "test": "bun test",