voiceio/pyproject.toml at main · Hugo0/voiceio · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
[build-system]
requires = ["setuptools>=68.0", "wheel"]
build-backend = "setuptools.build_meta"

[project]
name = "python-voiceio"
version = "1.15.0"
description = "Voice dictation for Linux. Speak → text, locally, instantly."
readme = "README.md"
license = "MIT"
requires-python = ">=3.11"
authors = [{ name = "Hugo Montenegro" }]
keywords = ["voice", "speech-to-text", "whisper", "linux", "dictation", "wayland", "ibus"]
classifiers = [
    "Development Status :: 4 - Beta",
    "Environment :: X11 Applications",
    "Intended Audience :: End Users/Desktop",
    "Operating System :: POSIX :: Linux",
    "Programming Language :: Python :: 3",
    "Topic :: Multimedia :: Sound/Audio :: Speech",
]
dependencies = [
    # Headless transcription only: decode, VAD, vocabulary ranking. Microphone,
    # hotkeys, tray and text injection live in the `desktop` extra so a server
    # install (`.[server]`) needs no display or audio stack.
    "faster-whisper>=1.0.0",
    "numpy>=1.24.0",
    "onnxruntime>=1.16.0",
    "wordfreq>=3.0",
]

[project.optional-dependencies]
# The dictation app: microphone capture, global hotkeys, tray icon. System
# packages (PortAudio, IBus, a C toolchain for evdev) are separate — see README.
desktop = [
    "sounddevice>=0.4.6",
    "evdev>=1.6.0; sys_platform == 'linux'",
    "pynput>=1.7.6",
    "pystray>=0.19",
    "Pillow>=10.0",
]
# Pre-1.0 name for the desktop install; kept so existing setups keep syncing.
linux = [
    "sounddevice>=0.4.6",
    "evdev>=1.6.0; sys_platform == 'linux'",
    "pynput>=1.7.6",
    "pystray>=0.19",
    "Pillow>=10.0",
]
# Headless HTTP/WebSocket server (`python -m voiceio.server`).
server = ["aiohttp>=3.9"]
# Text-to-speech: local (piper) and cloud (edge-tts) engines.
tts = ["piper-tts>=1.3.0", "edge-tts>=6.1.0"]
# `voiceio learn train`: LoRA fine-tuning (a GPU helps; see voiceio/learn/train.py)
train = ["torch>=2.2", "transformers>=4.40", "peft>=0.10"]
# Alternative decoder runtime (Parakeet & co.); see CONTRIBUTING.md "Model choice".
sherpa = ["sherpa-onnx>=1.12"]
dev = ["pytest>=7.0", "pytest-mock", "ruff>=0.15,<0.18"]

[project.urls]
Homepage = "https://voiceio.dev/"
Repository = "https://github.com/Hugo0/voiceio"
Issues = "https://github.com/Hugo0/voiceio/issues"
Changelog = "https://github.com/Hugo0/voiceio/releases"

[project.scripts]
voiceio = "voiceio.cli:_entry_point"
# Pre-1.0 GNOME shortcuts call this; kept through 1.x.
voiceio-toggle = "voiceio.cli:_cmd_toggle"

[tool.ruff]
# Explicit, so a new ruff release changing its defaults can't break CI.
lint.select = ["E4", "E7", "E9", "F"]
lint.per-file-ignores."voiceio/ibus/engine.py" = ["E402"]

[tool.setuptools.packages.find]
include = ["voiceio*"]

[tool.setuptools.package-data]
"voiceio.sounds" = ["*.wav"]
"voiceio.models" = ["*.onnx"]
"voiceio.data" = ["*.rules"]