voice-mode 4.2.0__tar.gz → 4.3.1__tar.gz

{voice_mode-4.2.0 → voice_mode-4.3.1}/CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [4.3.1] - 2025-09-03
+
+## [4.3.0] - 2025-09-03
+
 ## [4.2.0] - 2025-09-03
 
 ### Added
@@ -23,6 +27,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - Claude Code integration via stdin-receiver for hook-based audio
   - CLI command `play-sound` with theme, action, and sound selection
   - Enhances user experience with auditory feedback during operations
+  - MP3 support added for 90% file size reduction over WAV
+  - Recursive directory copying for complete sound font structure
+  - Three Bears sound fonts for baby-bear, mama-bear, and papa-bear agents
+  - Sound fonts disabled by default (VOICEMODE_SOUNDFONTS_ENABLED=false)
 
 - **🎭 Claude Code Deep Integration**
   - Extract and analyze Claude's conversation logs in real-time

{voice_mode-4.2.0 → voice_mode-4.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: voice-mode
-Version: 4.2.0
+Version: 4.3.1
 Summary: VoiceMode - Voice interaction capabilities for AI assistants (formerly voice-mcp)
 Project-URL: Homepage, https://github.com/mbailey/voicemode
 Project-URL: Repository, https://github.com/mbailey/voicemode

{voice_mode-4.2.0 → voice_mode-4.3.1}/voice_mode/__version__.py

@@ -1,3 +1,3 @@
 # This file is automatically updated by 'make release'
 # Do not edit manually
-__version__ = "4.2.0"
+__version__ = "4.3.1"

{voice_mode-4.2.0 → voice_mode-4.3.1}/voice_mode/cli.py

@@ -57,6 +57,14 @@ def voice_mode() -> None:
     voice_mode_main_cli()
 
 
+# Audio group for audio-related commands
+@voice_mode_main_cli.group()
+@click.help_option('-h', '--help', help='Show this message and exit')
+def audio():
+    """Audio transcription and playback commands."""
+    pass
+
+
 # Service group commands
 @voice_mode_main_cli.group()
 @click.help_option('-h', '--help', help='Show this message and exit')
@@ -1372,15 +1380,25 @@ cli.add_command(claude.claude_group)
 
 # Add exchanges to main CLI
 voice_mode_main_cli.add_command(exchanges_cmd.exchanges)
-voice_mode_main_cli.add_command(pronounce_commands.pronounce_group)
 voice_mode_main_cli.add_command(claude.claude_group)
 
-# Add transcribe to main CLI
-voice_mode_main_cli.add_command(transcribe_cmd.transcribe)
+# Note: We'll add these commands after the groups are defined
+# audio group will get transcribe and play commands
+# claude group will get hook command  
+# config group will get pronounce command
+
+
+# Now add the subcommands to their respective groups
+# Add transcribe command to audio group
+transcribe_audio_cmd = transcribe_cmd.transcribe.commands['audio']
+transcribe_audio_cmd.name = 'transcribe'
+audio.add_command(transcribe_audio_cmd)
 
-# Add hook command to main CLI
-voice_mode_main_cli.add_command(hook_cmd.hook)
+# Add hook command under claude group  
+claude.claude_group.add_command(hook_cmd.hook)
 
+# Add pronounce under config group
+config.add_command(pronounce_commands.pronounce_group)
 
 # Converse command - direct voice conversation from CLI
 @voice_mode_main_cli.command()
@@ -1759,7 +1777,7 @@ def update(force):
 
 
 # Sound Fonts command
-@voice_mode_main_cli.command("play-sound")
+@audio.command("play")
 @click.help_option('-h', '--help')
 @click.option('-t', '--tool', help='Tool name for direct command-line usage')
 @click.option('-a', '--action', default='start', type=click.Choice(['start', 'end']), help='Action type')

{voice_mode-4.2.0 → voice_mode-4.3.1}/voice_mode/cli_commands/hook.py

@@ -86,25 +86,32 @@ def stdin_receiver(tool_name, action, subagent_type, event, debug):
         print(f"[DEBUG] Processing: event={event_name}, tool={tool_name}, "
               f"action={action}, subagent={subagent_type}", file=sys.stderr)
     
-    # Find sound file using filesystem conventions
-    sound_file = find_sound_file(event_name, tool_name, subagent_type)
+    # Check if sound fonts are enabled
+    from voice_mode.config import SOUNDFONTS_ENABLED
     
-    if sound_file:
+    if not SOUNDFONTS_ENABLED:
         if debug:
-            print(f"[DEBUG] Found sound file: {sound_file}", file=sys.stderr)
-        
-        # Play the sound
-        player = Player()
-        success = player.play(str(sound_file))
-        
-        if debug:
-            if success:
-                print(f"[DEBUG] Sound played successfully", file=sys.stderr)
-            else:
-                print(f"[DEBUG] Failed to play sound", file=sys.stderr)
+            print(f"[DEBUG] Sound fonts are disabled (VOICEMODE_SOUNDFONTS_ENABLED=false)", file=sys.stderr)
     else:
-        if debug:
-            print(f"[DEBUG] No sound file found for this event", file=sys.stderr)
+        # Find sound file using filesystem conventions
+        sound_file = find_sound_file(event_name, tool_name, subagent_type)
+        
+        if sound_file:
+            if debug:
+                print(f"[DEBUG] Found sound file: {sound_file}", file=sys.stderr)
+            
+            # Play the sound
+            player = Player()
+            success = player.play(str(sound_file))
+            
+            if debug:
+                if success:
+                    print(f"[DEBUG] Sound played successfully", file=sys.stderr)
+                else:
+                    print(f"[DEBUG] Failed to play sound", file=sys.stderr)
+        else:
+            if debug:
+                print(f"[DEBUG] No sound file found for this event", file=sys.stderr)
     
     # Always exit 0 to not disrupt Claude Code
     sys.exit(0)
@@ -114,11 +121,11 @@ def find_sound_file(event: str, tool: str, subagent: Optional[str] = None) -> Op
     """
     Find sound file using filesystem conventions.
     
-    Tries paths in order:
-    1. Most specific: {event}/{tool}/subagent/{subagent}.wav (Task tool only)
-    2. Tool default: {event}/{tool}/default.wav
-    3. Event default: {event}/default.wav
-    4. Global fallback: fallback.wav
+    Tries paths in order (mp3 preferred over wav for size):
+    1. Most specific: {event}/{tool}/subagent/{subagent}.{mp3,wav} (Task tool only)
+    2. Tool default: {event}/{tool}/default.{mp3,wav}
+    3. Event default: {event}/default.{mp3,wav}
+    4. Global fallback: fallback.{mp3,wav}
     
     Args:
         event: Event name (PreToolUse, PostToolUse)
@@ -157,15 +164,20 @@ def find_sound_file(event: str, tool: str, subagent: Optional[str] = None) -> Op
     
     # 1. Most specific: subagent sound (Task tool only)
     if tool == 'task' and subagent:
+        # Try mp3 first (smaller), then wav
+        paths_to_try.append(base_path / event_dir / tool / 'subagent' / f'{subagent}.mp3')
         paths_to_try.append(base_path / event_dir / tool / 'subagent' / f'{subagent}.wav')
     
     # 2. Tool-specific default
+    paths_to_try.append(base_path / event_dir / tool / 'default.mp3')
     paths_to_try.append(base_path / event_dir / tool / 'default.wav')
     
     # 3. Event-level default
+    paths_to_try.append(base_path / event_dir / 'default.mp3')
     paths_to_try.append(base_path / event_dir / 'default.wav')
     
     # 4. Global fallback
+    paths_to_try.append(base_path / 'fallback.mp3')
     paths_to_try.append(base_path / 'fallback.wav')
     
     # Find first existing file

{voice_mode-4.2.0 → voice_mode-4.3.1}/voice_mode/config.py

@@ -101,6 +101,9 @@ def load_voicemode_env():
 # Enable audio feedback (true/false)
 # VOICEMODE_AUDIO_FEEDBACK=true
 
+# Enable sound fonts for tool use hooks (true/false)
+# VOICEMODE_SOUNDFONTS_ENABLED=false
+
 #############
 # Provider Configuration
 #############
@@ -371,6 +374,11 @@ FRONTEND_PORT = int(os.getenv("VOICEMODE_FRONTEND_PORT", "3000"))
 # Auto-enable services after installation
 SERVICE_AUTO_ENABLE = env_bool("VOICEMODE_SERVICE_AUTO_ENABLE", False)
 
+# ==================== SOUND FONTS CONFIGURATION ====================
+
+# Sound fonts are disabled by default to avoid annoying users with unexpected sounds
+SOUNDFONTS_ENABLED = env_bool("VOICEMODE_SOUNDFONTS_ENABLED", False)
+
 # ==================== AUDIO CONFIGURATION ====================
 
 # Audio parameters
@@ -555,6 +563,87 @@ def initialize_directories():
     # Create events log directory
     if EVENT_LOG_ENABLED:
         Path(EVENT_LOG_DIR).mkdir(parents=True, exist_ok=True)
+    
+    # Initialize sound fonts if not present
+    initialize_soundfonts()
+
+# ==================== SOUND FONTS INITIALIZATION ====================
+
+def initialize_soundfonts():
+    """Install default sound fonts from package data if not present."""
+    import shutil
+    import importlib.resources
+    
+    soundfonts_dir = BASE_DIR / "soundfonts"
+    default_soundfont_dir = soundfonts_dir / "default"
+    current_symlink = soundfonts_dir / "current"
+    
+    # Skip if soundfonts already exist (user has customized them)
+    if default_soundfont_dir.exists():
+        # Ensure symlink exists if directory exists
+        if not current_symlink.exists():
+            try:
+                current_symlink.symlink_to(default_soundfont_dir.resolve())
+            except OSError:
+                # Symlinks might not work on all systems
+                pass
+        return
+    
+    try:
+        # Create soundfonts directory
+        soundfonts_dir.mkdir(exist_ok=True)
+        
+        # Copy default soundfonts from package data
+        try:
+            # For Python 3.9+
+            from importlib.resources import files
+            package_soundfonts = files("voice_mode.data.soundfonts.default")
+            
+            if package_soundfonts.is_dir():
+                # Create the default directory
+                default_soundfont_dir.mkdir(exist_ok=True)
+                
+                # Recursively copy all files from package data
+                def copy_tree(src, dst):
+                    """Recursively copy directory tree from package data."""
+                    dst.mkdir(exist_ok=True)
+                    for item in src.iterdir():
+                        if item.is_file():
+                            target = dst / item.name
+                            target.write_bytes(item.read_bytes())
+                        elif item.is_dir():
+                            copy_tree(item, dst / item.name)
+                
+                # Copy entire tree structure
+                copy_tree(package_soundfonts, default_soundfont_dir)
+        except ImportError:
+            # Fallback for older Python versions
+            import pkg_resources
+            
+            # Create the default directory
+            default_soundfont_dir.mkdir(exist_ok=True)
+            
+            # List all resources in the soundfonts directory
+            resource_dir = "data/soundfonts/default"
+            if pkg_resources.resource_exists("voice_mode", resource_dir):
+                # This is a bit more complex with pkg_resources
+                # We'll need to manually copy the structure
+                pass
+        
+        # Create symlink to current soundfont (points to default)
+        if default_soundfont_dir.exists() and not current_symlink.exists():
+            try:
+                current_symlink.symlink_to(default_soundfont_dir.resolve())
+            except OSError:
+                # Symlinks might not work on all systems (e.g., Windows without admin)
+                pass
+                
+    except Exception as e:
+        # Don't fail initialization if soundfonts can't be installed
+        # They're optional and disabled by default
+        if DEBUG:
+            import logging
+            logging.getLogger("voicemode").debug(f"Could not initialize soundfonts: {e}")
 
 # ==================== UTILITY FUNCTIONS ====================