python-voiceio 0.3.10__tar.gz → 0.3.11__tar.gz

{python_voiceio-0.3.10/python_voiceio.egg-info → python_voiceio-0.3.11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: python-voiceio
-Version: 0.3.10
+Version: 0.3.11
 Summary: Speak → text, locally, instantly.
 Author: Hugo Montenegro
 License-Expression: MIT
@@ -26,16 +26,10 @@ Requires-Dist: numpy>=1.24.0
 Requires-Dist: onnxruntime>=1.16.0
 Requires-Dist: wordfreq>=3.0
 Requires-Dist: evdev>=1.6.0; sys_platform == "linux"
-Requires-Dist: pynput>=1.7.6; sys_platform == "win32"
-Requires-Dist: pynput>=1.7.6; sys_platform == "darwin"
+Requires-Dist: pynput>=1.7.6
 Requires-Dist: pyperclip>=1.8.0; sys_platform == "win32"
 Requires-Dist: win11toast>=0.36; sys_platform == "win32"
-Provides-Extra: x11
-Requires-Dist: pynput>=1.7.6; extra == "x11"
-Provides-Extra: mac
-Requires-Dist: pynput>=1.7.6; extra == "mac"
 Provides-Extra: win
-Requires-Dist: pynput>=1.7.6; extra == "win"
 Requires-Dist: pyperclip>=1.8.0; extra == "win"
 Requires-Dist: win11toast>=0.36; extra == "win"
 Provides-Extra: tray
@@ -52,6 +46,15 @@ Dynamic: license-file
 
 # voiceio
 
+```
+ ██╗   ██╗ ██████╗ ██╗ ██████╗███████╗██╗ ██████╗
+ ██║   ██║██╔═══██╗██║██╔════╝██╔════╝██║██╔═══██╗
+ ██║   ██║██║   ██║██║██║     █████╗  ██║██║   ██║
+ ╚██╗ ██╔╝██║   ██║██║██║     ██╔══╝  ██║██║   ██║
+  ╚████╔╝ ╚██████╔╝██║╚██████╗███████╗██║╚██████╔╝
+   ╚═══╝   ╚═════╝ ╚═╝ ╚═════╝╚══════╝╚═╝ ╚═════╝
+```
+
 [![CI](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml/badge.svg)](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml)
 [![PyPI](https://img.shields.io/pypi/v/python-voiceio)](https://pypi.org/project/python-voiceio/)
 [![Python](https://img.shields.io/pypi/pyversions/python-voiceio)](https://pypi.org/project/python-voiceio/)
@@ -60,11 +63,13 @@ Dynamic: license-file
 
 Speak → text, locally, instantly.
 
+https://github.com/user-attachments/assets/9cf5d1ac-b4bb-4cf8-b775-7a66dc16b376
+
 ## Quick start
 
 ```bash
 # 1. Install system dependencies (Ubuntu/Debian)
-sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi portaudio19-dev
+sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi python3-dev portaudio19-dev
 
 # 2. Install voiceio
 pipx install python-voiceio
@@ -79,7 +84,7 @@ That's it. Press **Ctrl+Alt+V** (or your chosen hotkey) to start dictating.
 <summary><strong>Fedora</strong></summary>
 
 ```bash
-sudo dnf install pipx ibus python3-gobject portaudio-devel
+sudo dnf install pipx ibus python3-gobject python3-devel portaudio-devel
 pipx install python-voiceio
 voiceio setup
 ```
@@ -90,6 +95,7 @@ voiceio setup
 
 ```bash
 sudo pacman -S python-pipx ibus python-gobject portaudio
+# Note: Arch includes Python headers by default with the python package
 pipx install python-voiceio
 voiceio setup
 ```

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/README.md

@@ -1,5 +1,14 @@
 # voiceio
 
+```
+ ██╗   ██╗ ██████╗ ██╗ ██████╗███████╗██╗ ██████╗
+ ██║   ██║██╔═══██╗██║██╔════╝██╔════╝██║██╔═══██╗
+ ██║   ██║██║   ██║██║██║     █████╗  ██║██║   ██║
+ ╚██╗ ██╔╝██║   ██║██║██║     ██╔══╝  ██║██║   ██║
+  ╚████╔╝ ╚██████╔╝██║╚██████╗███████╗██║╚██████╔╝
+   ╚═══╝   ╚═════╝ ╚═╝ ╚═════╝╚══════╝╚═╝ ╚═════╝
+```
+
 [![CI](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml/badge.svg)](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml)
 [![PyPI](https://img.shields.io/pypi/v/python-voiceio)](https://pypi.org/project/python-voiceio/)
 [![Python](https://img.shields.io/pypi/pyversions/python-voiceio)](https://pypi.org/project/python-voiceio/)
@@ -8,11 +17,13 @@
 
 Speak → text, locally, instantly.
 
+https://github.com/user-attachments/assets/9cf5d1ac-b4bb-4cf8-b775-7a66dc16b376
+
 ## Quick start
 
 ```bash
 # 1. Install system dependencies (Ubuntu/Debian)
-sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi portaudio19-dev
+sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi python3-dev portaudio19-dev
 
 # 2. Install voiceio
 pipx install python-voiceio
@@ -27,7 +38,7 @@ That's it. Press **Ctrl+Alt+V** (or your chosen hotkey) to start dictating.
 <summary><strong>Fedora</strong></summary>
 
 ```bash
-sudo dnf install pipx ibus python3-gobject portaudio-devel
+sudo dnf install pipx ibus python3-gobject python3-devel portaudio-devel
 pipx install python-voiceio
 voiceio setup
 ```
@@ -38,6 +49,7 @@ voiceio setup
 
 ```bash
 sudo pacman -S python-pipx ibus python-gobject portaudio
+# Note: Arch includes Python headers by default with the python package
 pipx install python-voiceio
 voiceio setup
 ```

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "python-voiceio"
-version = "0.3.10"
+version = "0.3.11"
 description = "Speak → text, locally, instantly."
 readme = "README.md"
 license = "MIT"
@@ -28,16 +28,13 @@ dependencies = [
     "onnxruntime>=1.16.0",
     "wordfreq>=3.0",
     "evdev>=1.6.0; sys_platform == 'linux'",
-    "pynput>=1.7.6; sys_platform == 'win32'",
-    "pynput>=1.7.6; sys_platform == 'darwin'",
+    "pynput>=1.7.6",
     "pyperclip>=1.8.0; sys_platform == 'win32'",
     "win11toast>=0.36; sys_platform == 'win32'",
 ]
 
 [project.optional-dependencies]
-x11 = ["pynput>=1.7.6"]
-mac = ["pynput>=1.7.6"]
-win = ["pynput>=1.7.6", "pyperclip>=1.8.0", "win11toast>=0.36"]
+win = ["pyperclip>=1.8.0", "win11toast>=0.36"]
 tray = ["pystray>=0.19", "Pillow>=10.0"]
 tts = ["piper-tts>=1.2.0"]
 tts-cloud = ["edge-tts>=6.1.0"]

{python_voiceio-0.3.10 → python_voiceio-0.3.11/python_voiceio.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: python-voiceio
-Version: 0.3.10
+Version: 0.3.11
 Summary: Speak → text, locally, instantly.
 Author: Hugo Montenegro
 License-Expression: MIT
@@ -26,16 +26,10 @@ Requires-Dist: numpy>=1.24.0
 Requires-Dist: onnxruntime>=1.16.0
 Requires-Dist: wordfreq>=3.0
 Requires-Dist: evdev>=1.6.0; sys_platform == "linux"
-Requires-Dist: pynput>=1.7.6; sys_platform == "win32"
-Requires-Dist: pynput>=1.7.6; sys_platform == "darwin"
+Requires-Dist: pynput>=1.7.6
 Requires-Dist: pyperclip>=1.8.0; sys_platform == "win32"
 Requires-Dist: win11toast>=0.36; sys_platform == "win32"
-Provides-Extra: x11
-Requires-Dist: pynput>=1.7.6; extra == "x11"
-Provides-Extra: mac
-Requires-Dist: pynput>=1.7.6; extra == "mac"
 Provides-Extra: win
-Requires-Dist: pynput>=1.7.6; extra == "win"
 Requires-Dist: pyperclip>=1.8.0; extra == "win"
 Requires-Dist: win11toast>=0.36; extra == "win"
 Provides-Extra: tray
@@ -52,6 +46,15 @@ Dynamic: license-file
 
 # voiceio
 
+```
+ ██╗   ██╗ ██████╗ ██╗ ██████╗███████╗██╗ ██████╗
+ ██║   ██║██╔═══██╗██║██╔════╝██╔════╝██║██╔═══██╗
+ ██║   ██║██║   ██║██║██║     █████╗  ██║██║   ██║
+ ╚██╗ ██╔╝██║   ██║██║██║     ██╔══╝  ██║██║   ██║
+  ╚████╔╝ ╚██████╔╝██║╚██████╗███████╗██║╚██████╔╝
+   ╚═══╝   ╚═════╝ ╚═╝ ╚═════╝╚══════╝╚═╝ ╚═════╝
+```
+
 [![CI](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml/badge.svg)](https://github.com/Hugo0/voiceio/actions/workflows/ci.yml)
 [![PyPI](https://img.shields.io/pypi/v/python-voiceio)](https://pypi.org/project/python-voiceio/)
 [![Python](https://img.shields.io/pypi/pyversions/python-voiceio)](https://pypi.org/project/python-voiceio/)
@@ -60,11 +63,13 @@ Dynamic: license-file
 
 Speak → text, locally, instantly.
 
+https://github.com/user-attachments/assets/9cf5d1ac-b4bb-4cf8-b775-7a66dc16b376
+
 ## Quick start
 
 ```bash
 # 1. Install system dependencies (Ubuntu/Debian)
-sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi portaudio19-dev
+sudo apt install pipx ibus gir1.2-ibus-1.0 python3-gi python3-dev portaudio19-dev
 
 # 2. Install voiceio
 pipx install python-voiceio
@@ -79,7 +84,7 @@ That's it. Press **Ctrl+Alt+V** (or your chosen hotkey) to start dictating.
 <summary><strong>Fedora</strong></summary>
 
 ```bash
-sudo dnf install pipx ibus python3-gobject portaudio-devel
+sudo dnf install pipx ibus python3-gobject python3-devel portaudio-devel
 pipx install python-voiceio
 voiceio setup
 ```
@@ -90,6 +95,7 @@ voiceio setup
 
 ```bash
 sudo pacman -S python-pipx ibus python-gobject portaudio
+# Note: Arch includes Python headers by default with the python package
 pipx install python-voiceio
 voiceio setup
 ```

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/python_voiceio.egg-info/requires.txt

@@ -3,15 +3,12 @@ sounddevice>=0.4.6
 numpy>=1.24.0
 onnxruntime>=1.16.0
 wordfreq>=3.0
-
-[:sys_platform == "darwin"]
 pynput>=1.7.6
 
 [:sys_platform == "linux"]
 evdev>=1.6.0
 
 [:sys_platform == "win32"]
-pynput>=1.7.6
 pyperclip>=1.8.0
 win11toast>=0.36
 
@@ -19,9 +16,6 @@ win11toast>=0.36
 pytest>=7.0
 pytest-mock
 
-[mac]
-pynput>=1.7.6
-
 [tray]
 pystray>=0.19
 Pillow>=10.0
@@ -33,9 +27,5 @@ piper-tts>=1.2.0
 edge-tts>=6.1.0
 
 [win]
-pynput>=1.7.6
 pyperclip>=1.8.0
 win11toast>=0.36
-
-[x11]
-pynput>=1.7.6

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/tests/test_commands.py

@@ -59,30 +59,36 @@ class TestFormatting:
 
 class TestUndo:
     def test_scratch_that(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("hello world scratch that")
         assert result == "hello world"
         assert cp.undo_requested is True
 
     def test_undo_that(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("hello world undo that")
         assert result == "hello world"
         assert cp.undo_requested is True
 
     def test_scratch_that_only(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("scratch that")
         assert result == ""
         assert cp.undo_requested is True
 
     def test_undo_resets_between_calls(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         cp.process("scratch that")
         assert cp.undo_requested is True
         cp.process("hello world")
         assert cp.undo_requested is False
 
+    def test_editing_off_ignores_scratch(self):
+        cp = CommandProcessor(editing=False)
+        result = cp.process("hello world scratch that")
+        assert result == "hello world scratch that"
+        assert cp.undo_requested is False
+
 
 class TestEdgeCases:
     def test_no_commands(self):
@@ -127,27 +133,27 @@ class TestEdgeCases:
 
 class TestCorrectThat:
     def test_correct_that_sets_flag(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("hello world correct that")
         assert cp.flag_requested is True
         assert cp.flagged_word == "world"
         assert result == "hello"
 
     def test_correct_that_no_preceding_word(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("correct that")
         assert cp.flag_requested is True
         assert cp.flagged_word == ""
         assert result == ""
 
     def test_correct_that_returns_text_before(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         result = cp.process("one two three correct that")
         assert result == "one two"
         assert cp.flagged_word == "three"
 
     def test_flag_resets_between_calls(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         cp.process("hello correct that")
         assert cp.flag_requested is True
         cp.process("hello world")
@@ -155,11 +161,17 @@ class TestCorrectThat:
         assert cp.flagged_word == ""
 
     def test_undo_not_set_on_flag(self):
-        cp = CommandProcessor()
+        cp = CommandProcessor(editing=True)
         cp.process("hello correct that")
         assert cp.flag_requested is True
         assert cp.undo_requested is False
 
+    def test_editing_off_ignores_correct(self):
+        cp = CommandProcessor(editing=False)
+        result = cp.process("hello world correct that")
+        assert result == "hello world correct that"
+        assert cp.flag_requested is False
+
 
 class TestSpacing:
     def test_no_double_spaces(self):

python_voiceio-0.3.11/voiceio/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.3.11"

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/app.py

@@ -126,6 +126,24 @@ class VoiceIO:
         self._socket: SocketHotkey | None = None
         if self._hotkey.name != "socket":
             self._socket = SocketHotkey()
+        elif self.platform.desktop not in ("GNOME", "KDE"):
+            # Socket backend has no physical hotkey listener — user must bind
+            # `voiceio toggle` to a key in their WM config.
+            log.warning(
+                "Hotkey backend is 'socket' on desktop '%s'. "
+                "Physical hotkey capture is not available. "
+                "Bind 'voiceio toggle' to a key in your window manager config "
+                "(e.g. i3: bindsym Ctrl+Alt+v exec voiceio toggle). "
+                "Or: sudo usermod -aG input $USER && log out/in to enable evdev.",
+                self.platform.desktop,
+            )
+            print(
+                f"\n  NOTE: No physical hotkey listener available on '{self.platform.desktop}'.\n"
+                f"  Bind 'voiceio toggle' to a key in your WM config, e.g.:\n"
+                f"    i3/sway:  bindsym Ctrl+Alt+v exec voiceio toggle\n"
+                f"    hyprland: bind = CTRL ALT, V, exec, voiceio toggle\n"
+                f"  Or fix evdev: sudo usermod -aG input $USER (then log out/in)\n",
+            )
 
         print(f"Loading model '{cfg.model.name}'...", end="", flush=True)
         t0 = time.monotonic()
@@ -146,7 +164,7 @@ class VoiceIO:
         if vocab:
             self.transcriber.set_initial_prompt(vocab)
 
-        self._command_processor = CommandProcessor(enabled=cfg.commands.enabled)
+        self._command_processor = CommandProcessor(enabled=cfg.commands.enabled, editing=cfg.commands.editing)
         self._cleanup = cfg.output.punctuation_cleanup
         self._number_conversion = cfg.output.number_conversion
         self._streaming = cfg.output.streaming

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/commands.py

@@ -83,9 +83,12 @@ def _normalize_spacing(text: str) -> str:
 class CommandProcessor:
     """Detects and replaces voice commands in transcribed text."""
 
-    def __init__(self, enabled: bool = True):
+    def __init__(self, enabled: bool = True, editing: bool = False):
         self._enabled = enabled
-        self._commands = DEFAULT_COMMANDS
+        self._commands = {
+            k: v for k, v in DEFAULT_COMMANDS.items()
+            if editing or v not in (_UNDO_SENTINEL, _FLAG_SENTINEL)
+        }
         self._max_words = max(len(k) for k in self._commands) if self._commands else 1
         self.undo_requested = False
         self.flag_requested = False

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/config.py

@@ -85,6 +85,7 @@ class DaemonConfig:
 @dataclass
 class CommandsConfig:
     enabled: bool = True
+    editing: bool = False
 
 
 @dataclass

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/platform.py

@@ -234,6 +234,10 @@ def open_in_terminal(cmd: list[str]) -> bool:
     _TERMINALS = [
         (["gnome-terminal", "--"], "gnome-terminal"),
         (["konsole", "-e"], "konsole"),
+        (["alacritty", "-e"], "alacritty"),
+        (["kitty", "--"], "kitty"),
+        (["foot", "--"], "foot"),
+        (["wezterm", "start", "--"], "wezterm"),
         (["xfce4-terminal", "-e"], "xfce4-terminal"),
         (["xterm", "-e"], "xterm"),
     ]

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/typers/clipboard.py

@@ -80,7 +80,7 @@ class ClipboardTyper:
             return ProbeResult(
                 ok=False,
                 reason="No clipboard tool found",
-                fix_hint="Install xclip (X11), wl-copy (Wayland), or pbcopy (macOS).",
+                fix_hint="Install xclip + xdotool (X11), wl-copy + ydotool/wtype (Wayland), or pbcopy (macOS).",
             )
         if sys.platform == "win32":
             try:

{python_voiceio-0.3.10 → python_voiceio-0.3.11}/voiceio/wizard.py

@@ -945,10 +945,11 @@ def run_wizard() -> None:
                  "will use GPU" if checks["cuda"] else "will use CPU (still fast)",
                  optional=True)
 
-    if checks["display"] == "wayland":
+    if checks["is_linux"]:
         _print_check("Input group (evdev)", checks["input_group"],
-                     "" if checks["input_group"] else "optional: sudo usermod -aG input $USER",
-                     optional=True)
+                     "evdev hotkeys available" if checks["input_group"]
+                     else "sudo usermod -aG input $USER (then log out/in)",
+                     optional=checks["input_group"])
 
     # Tray icon
     from voiceio.tray import probe_availability
@@ -1067,13 +1068,12 @@ def run_wizard() -> None:
             else:
                 print(f"  {YELLOW}⚠{RESET}  {DIM}Could not install IBus component, will use fallback{RESET}")
 
-        if checks["display"] == "wayland":
-            if checks["input_group"]:
-                backend = "evdev"
-            else:
-                backend = "socket"
+        if checks["input_group"]:
+            backend = "evdev"
+        elif checks["pynput"] and checks["display"] != "wayland":
+            backend = "auto"  # pynput works on X11
         else:
-            backend = "auto"
+            backend = "socket"
 
     # ── Defaults for optional settings ──────────────────────────────────
     sound_enabled = True
@@ -1286,9 +1286,12 @@ def run_wizard() -> None:
                 print(f"  {YELLOW}⚠{RESET}  Auto-setup failed. Add manually in Settings → Keyboard → Shortcuts:")
                 print("    Command: voiceio-toggle")
         elif backend == "socket":
-            print(f"\n  {YELLOW}ℹ{RESET}  Add a keyboard shortcut manually in your DE settings:")
-            print(f"    Shortcut: {BOLD}{hotkey}{RESET}")
-            print(f"    Command:  {BOLD}voiceio-toggle{RESET}")
+            print(f"\n  {YELLOW}ℹ{RESET}  Add a keyboard shortcut in your window manager config:")
+            print(f"    {DIM}i3/sway:{RESET}   bindsym Ctrl+Alt+v exec voiceio toggle")
+            print(f"    {DIM}hyprland:{RESET}  bind = CTRL ALT, V, exec, voiceio toggle")
+            print(f"    {DIM}other:{RESET}     bind {BOLD}{hotkey}{RESET} → {BOLD}voiceio toggle{RESET}")
+            print(f"\n    {DIM}Or enable evdev (no WM config needed):{RESET}")
+            print(f"    sudo usermod -aG input $USER  {DIM}(then log out and back in){RESET}")
 
     # Autostart
     from voiceio.service import install_service

python_voiceio-0.3.10/voiceio/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.3.10"