voice-mode 3.34.3__tar.gz → 4.0.1__tar.gz

{voice_mode-3.34.3 → voice_mode-4.0.1}/.gitignore

@@ -117,3 +117,9 @@ models/
 *.mlpackage/
 *.mlmodel
 *.mlmodelc/
+
+# Coverage reports
+htmlcov/
+.coverage
+.coverage.*
+coverage.xml

{voice_mode-3.34.3 → voice_mode-4.0.1}/CHANGELOG.md

@@ -7,6 +7,49 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [4.0.1] - 2025-09-01
+
+### Removed
+- Removed `whisperx` optional dependency to fix PyPI upload compatibility
+  - The dependency was specified as a Git URL which is not allowed for PyPI packages
+  - WhisperX functionality was recently added and not essential for core features
+
+## [4.0.0] - 2025-08-31
+
+### BREAKING CHANGES
+- **Unified voice configuration system**
+  - **BREAKING**: Replaced `.voices.txt` files with unified `.voicemode.env` configuration
+  - Changed environment variable from `VOICEMODE_TTS_VOICES` to `VOICEMODE_VOICES` for simplicity
+  - Implemented cascading configuration: env vars > project configs > global config  
+  - Added directory tree walking for project-specific configuration discovery
+  - Supports runtime configuration reloading via MCP tools
+  - **Migration Required**: Users must migrate from `.voices.txt` to `.voicemode.env` with `VOICEMODE_VOICES=voice1,voice2` format
+
+### Added
+
+- **Comprehensive test coverage reporting system**
+  - Integration with pytest-cov for coverage measurement
+  - HTML coverage reports generated in htmlcov/ directory  
+  - Coverage badges and metrics for monitoring code quality
+  - Automated coverage reporting in CI/CD pipeline
+  
+- **Word-level timestamps for transcription**
+  - Enhanced transcription output with word-level timing information
+  - Support for SubRip (SRT) format output with precise word timestamps
+  - New transcription CLI command for processing audio files
+  - Comprehensive transcription backend supporting multiple formats
+  - Word timing data available for improved accessibility and analysis
+
+- **Enhanced voice selection guide**
+  - Comprehensive documentation for voice selection across different providers
+  - Clear migration instructions from old `.voices.txt` system
+  
+### Removed  
+- **Legacy voice preference system**
+  - Removed 578 lines of old `voice_preferences.py` system
+  - Eliminated unreliable `.voices.txt` file parsing
+  - Removed associated test files for deprecated voice preference system
+
 ## [3.34.3] - 2025-08-26
 
 ### Changed

{voice_mode-3.34.3 → voice_mode-4.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: voice-mode
-Version: 3.34.3
+Version: 4.0.1
 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
@@ -66,9 +66,12 @@ Requires-Dist: pandas>=2.0.0; extra == 'notebooks'
 Provides-Extra: scripts
 Requires-Dist: flask>=3.0.0; extra == 'scripts'
 Provides-Extra: test
+Requires-Dist: coverage[toml]>=7.4.0; extra == 'test'
 Requires-Dist: pytest-asyncio>=0.21.0; extra == 'test'
-Requires-Dist: pytest-cov>=4.0.0; extra == 'test'
+Requires-Dist: pytest-cov>=4.1.0; extra == 'test'
 Requires-Dist: pytest-mock>=3.10.0; extra == 'test'
+Requires-Dist: pytest-timeout>=2.2.0; extra == 'test'
+Requires-Dist: pytest-xdist>=3.5.0; extra == 'test'
 Requires-Dist: pytest>=7.0.0; extra == 'test'
 Description-Content-Type: text/markdown
 

{voice_mode-3.34.3 → voice_mode-4.0.1}/pyproject.toml

@@ -67,8 +67,11 @@ dev = [
 test = [
     "pytest>=7.0.0",
     "pytest-asyncio>=0.21.0",
-    "pytest-cov>=4.0.0",
+    "pytest-cov>=4.1.0",
     "pytest-mock>=3.10.0",
+    "pytest-xdist>=3.5.0",  # For parallel testing
+    "pytest-timeout>=2.2.0",  # For test timeouts
+    "coverage[toml]>=7.4.0",
 ]
 notebooks = [
     "gradio>=4.0.0",
@@ -144,10 +147,65 @@ exclude = [
 [tool.hatch.version]
 path = "voice_mode/__version__.py"
 
+[tool.hatch.metadata]
+allow-direct-references = true
+
 [tool.pytest.ini_options]
+minversion = "7.0"
 testpaths = ["tests"]
 python_files = "test_*.py"
 python_classes = "Test*"
 python_functions = "test_*"
-# Exclude manual test directory
-addopts = "--ignore=tests/manual"
+asyncio_mode = "auto"
+addopts = [
+    "-ra",
+    "--strict-markers",
+    "--strict-config",
+    "--ignore=tests/manual",
+    "--cov=voice_mode",
+    "--cov-branch",
+    "--cov-report=term-missing:skip-covered",
+    "--cov-report=html",
+    "--cov-report=xml",
+]
+markers = [
+    "unit: Unit tests (fast, isolated)",
+    "integration: Integration tests (may interact with services)",
+    "slow: Tests that take > 1s",
+    "manual: Manual tests requiring human interaction",
+]
+filterwarnings = [
+    "ignore::DeprecationWarning",
+]
+
+[tool.coverage.run]
+source = ["voice_mode"]
+branch = true
+parallel = true
+omit = [
+    "*/tests/*",
+    "*/test_*.py",
+    "*/__pycache__/*",
+    "*/site-packages/*",
+    "test-env/*",
+]
+
+[tool.coverage.report]
+exclude_lines = [
+    "pragma: no cover",
+    "def __repr__",
+    "if TYPE_CHECKING:",
+    "raise NotImplementedError",
+    "if __name__ == .__main__.:",
+    "@abstractmethod",
+    "except ImportError:",
+]
+precision = 2
+skip_covered = true
+show_missing = true
+
+[tool.coverage.html]
+directory = "htmlcov"
+
+[tool.coverage.xml]
+output = "coverage.xml"

{voice_mode-3.34.3 → voice_mode-4.0.1}/voice_mode/__version__.py

@@ -1,3 +1,3 @@
 # This file is automatically updated by 'make release'
 # Do not edit manually
-__version__ = "3.34.3"
+__version__ = "4.0.1"

{voice_mode-3.34.3 → voice_mode-4.0.1}/voice_mode/cli.py

@@ -1359,13 +1359,18 @@ def cli():
 
 # Import subcommand groups
 from voice_mode.cli_commands import exchanges as exchanges_cmd
+from voice_mode.cli_commands import transcribe as transcribe_cmd
 
 # Add subcommands to legacy CLI
 cli.add_command(exchanges_cmd.exchanges)
+cli.add_command(transcribe_cmd.transcribe)
 
 # Add exchanges to main CLI
 voice_mode_main_cli.add_command(exchanges_cmd.exchanges)
 
+# Add transcribe to main CLI
+voice_mode_main_cli.add_command(transcribe_cmd.transcribe)
+
 
 # Converse command - direct voice conversation from CLI
 @voice_mode_main_cli.command()

voice_mode-4.0.1/voice_mode/cli_commands/transcribe.py

@@ -0,0 +1,141 @@
+"""CLI command for audio transcription."""
+
+import click
+import json
+import asyncio
+from pathlib import Path
+from typing import Optional
+
+from voice_mode.tools.transcription import (
+    transcribe_audio,
+    TranscriptionBackend,
+    OutputFormat
+)
+
+
+@click.group()
+def transcribe():
+    """Audio transcription with word-level timestamps."""
+    pass
+
+
+@transcribe.command("audio")
+@click.argument('audio_file', type=click.Path(exists=True))
+@click.option('--words', is_flag=True, help='Include word-level timestamps')
+@click.option(
+    '--backend',
+    type=click.Choice(['openai', 'whisperx', 'whisper-cpp']),
+    default='openai',
+    help='Transcription backend to use'
+)
+@click.option(
+    '--format',
+    'output_format',
+    type=click.Choice(['json', 'srt', 'vtt', 'csv']),
+    default='json',
+    help='Output format for transcription'
+)
+@click.option('--output', '-o', type=click.Path(), help='Save transcription to file')
+@click.option('--language', help='Language code (e.g., en, es, fr)')
+@click.option('--model', default='whisper-1', help='Model to use (for OpenAI backend)')
+def audio_command(
+    audio_file: str,
+    words: bool,
+    backend: str,
+    output_format: str,
+    output: Optional[str],
+    language: Optional[str],
+    model: str
+):
+    """
+    Transcribe audio with optional word-level timestamps.
+    
+    Examples:
+    
+        voice-mode transcribe audio recording.mp3
+        
+        voice-mode transcribe audio interview.wav --words
+        
+        voice-mode transcribe audio podcast.mp3 --words --format srt -o subtitles.srt
+        
+        voice-mode transcribe audio spanish.mp3 --language es --backend whisperx
+    """
+    async def run():
+        # Perform transcription
+        result = await transcribe_audio(
+            audio_file=audio_file,
+            word_timestamps=words,
+            backend=TranscriptionBackend(backend),
+            output_format=OutputFormat(output_format),
+            language=language,
+            model=model
+        )
+        
+        # Check for errors
+        if not result.get("success", False):
+            error_msg = result.get("error", "Unknown error occurred")
+            click.echo(f"Error: {error_msg}", err=True)
+            return
+        
+        # Format output
+        if output_format == 'json':
+            # Remove internal fields for cleaner output
+            output_result = {k: v for k, v in result.items() 
+                           if k not in ['formatted_content']}
+            content = json.dumps(output_result, indent=2)
+        elif "formatted_content" in result:
+            content = result["formatted_content"]
+        else:
+            # Fallback to JSON if format conversion failed
+            content = json.dumps(result, indent=2)
+        
+        # Write output
+        if output:
+            Path(output).write_text(content)
+            click.echo(f"Transcription saved to {output}")
+        else:
+            click.echo(content)
+    
+    # Run async function
+    asyncio.run(run())
+
+
+# For backward compatibility, also provide a direct command
+@click.command('transcribe-audio')
+@click.argument('audio_file', type=click.Path(exists=True))
+@click.option('--words', is_flag=True, help='Include word-level timestamps')
+@click.option(
+    '--backend',
+    type=click.Choice(['openai', 'whisperx', 'whisper-cpp']),
+    default='openai',
+    help='Transcription backend'
+)
+@click.option(
+    '--format',
+    'output_format',
+    type=click.Choice(['json', 'srt', 'vtt', 'csv']),
+    default='json',
+    help='Output format'
+)
+@click.option('--output', '-o', type=click.Path(), help='Save to file')
+@click.option('--language', help='Language code')
+@click.option('--model', default='whisper-1', help='Model to use')
+def transcribe_audio_command(
+    audio_file: str,
+    words: bool,
+    backend: str,
+    output_format: str,
+    output: Optional[str],
+    language: Optional[str],
+    model: str
+):
+    """Direct transcription command for backward compatibility."""
+    audio_command.callback(
+        audio_file=audio_file,
+        words=words,
+        backend=backend,
+        output_format=output_format,
+        output=output,
+        language=language,
+        model=model
+    )

{voice_mode-3.34.3 → voice_mode-4.0.1}/voice_mode/config.py

@@ -15,21 +15,66 @@ from datetime import datetime
 
 # ==================== ENVIRONMENT CONFIGURATION ====================
 
+def find_voicemode_env_files() -> list[Path]:
+    """
+    Find .voicemode.env files by walking up the directory tree.
+    
+    Looks for (in order of priority - closest to current directory wins):
+    1. .voicemode.env in current or parent directories  
+    2. .voicemode/voicemode.env in current or parent directories
+    3. ~/.voicemode/voicemode.env in user home (global config)
+    
+    Returns:
+        List of Path objects in loading order (global first, then project-specific)
+    """
+    config_files = []
+    
+    # First add global config (lowest priority - loaded first)
+    global_config = Path.home() / ".voicemode" / "voicemode.env"
+    
+    # Backwards compatibility: check for old filename
+    if not global_config.exists():
+        old_global = Path.home() / ".voicemode" / ".voicemode.env"
+        if old_global.exists():
+            global_config = old_global
+    
+    if global_config.exists():
+        config_files.append(global_config)
+    
+    # Then walk up directory tree for project-specific configs (higher priority)
+    current_dir = Path.cwd()
+    project_configs = []
+    
+    while current_dir != current_dir.parent:
+        # Check for standalone .voicemode.env first
+        standalone_file = current_dir / ".voicemode.env"
+        if standalone_file.exists():
+            project_configs.append(standalone_file)
+            break  # Stop at first found (closest wins)
+            
+        # Then check .voicemode/voicemode.env  
+        dir_file = current_dir / ".voicemode" / "voicemode.env"
+        # Skip if this is the global config file (already added)
+        if dir_file.exists() and dir_file != global_config:
+            project_configs.append(dir_file)
+            break  # Stop at first found (closest wins)
+        
+        current_dir = current_dir.parent
+    
+    # Add project configs (they were collected closest-first, so add as-is)
+    config_files.extend(project_configs)
+    
+    return config_files
+
+
 def load_voicemode_env():
-    """Load configuration from voicemode.env file if it exists, creating a default if not."""
-    # Try new filename first
-    config_path = Path.home() / ".voicemode" / "voicemode.env"
-    
-    # Backwards compatibility: check for old filename if new doesn't exist
-    if not config_path.exists():
-        old_path = Path.home() / ".voicemode" / ".voicemode.env"
-        if old_path.exists():
-            config_path = old_path
-            print(f"Warning: Using deprecated .voicemode.env - please rename to voicemode.env")
-    
-    if not config_path.exists():
-        # Create default template
-        config_path.parent.mkdir(parents=True, exist_ok=True)
+    """Load configuration from voicemode.env files, with cascading from global to project-specific."""
+    config_files = find_voicemode_env_files()
+    
+    # If no config files found, create default global config
+    if not config_files:
+        default_path = Path.home() / ".voicemode" / "voicemode.env"
+        default_path.parent.mkdir(parents=True, exist_ok=True)
         default_config = '''# Voice Mode Configuration File
 # This file is automatically generated and can be customized
 # Environment variables always take precedence over this file
@@ -66,8 +111,8 @@ def load_voicemode_env():
 # Comma-separated list of STT endpoints
 # VOICEMODE_STT_BASE_URLS=http://127.0.0.1:2022/v1,https://api.openai.com/v1
 
-# Comma-separated list of preferred voices
-# VOICEMODE_TTS_VOICES=af_sky,alloy
+# Comma-separated list of preferred voices  
+# VOICEMODE_VOICES=af_sky,alloy
 
 # Comma-separated list of preferred models
 # VOICEMODE_TTS_MODELS=tts-1,tts-1-hd,gpt-4o-mini-tts
@@ -127,26 +172,28 @@ def load_voicemode_env():
 # LIVEKIT_API_KEY=devkey
 # LIVEKIT_API_SECRET=secret
 '''
-        with open(config_path, 'w') as f:
+        with open(default_path, 'w') as f:
             f.write(default_config)
-        os.chmod(config_path, 0o600)  # Secure permissions
-    
-    # Load configuration from file
-    if config_path.exists():
-        with open(config_path, 'r') as f:
-            for line in f:
-                line = line.strip()
-                # Skip comments and empty lines
-                if not line or line.startswith('#'):
-                    continue
-                # Parse KEY=VALUE format
-                if '=' in line:
-                    key, value = line.split('=', 1)
-                    key = key.strip()
-                    value = value.strip()
-                    # Only set if not already in environment
-                    if key and key not in os.environ:
-                        os.environ[key] = value
+        os.chmod(default_path, 0o600)  # Secure permissions
+        config_files = [default_path]
+    
+    # Load configuration from all files in order (global first, project-specific last)
+    for config_path in config_files:
+        if config_path.exists():
+            with open(config_path, 'r') as f:
+                for line in f:
+                    line = line.strip()
+                    # Skip comments and empty lines
+                    if not line or line.startswith('#'):
+                        continue
+                    # Parse KEY=VALUE format
+                    if '=' in line:
+                        key, value = line.split('=', 1)
+                        key = key.strip()
+                        value = value.strip()
+                        # Only set if not already in environment (env vars take precedence)
+                        if key and key not in os.environ:
+                            os.environ[key] = value
 
 # Load configuration file before other configuration
 load_voicemode_env()
@@ -222,13 +269,68 @@ def parse_comma_list(env_var: str, fallback: str) -> list:
 # New provider endpoint lists configuration
 TTS_BASE_URLS = parse_comma_list("VOICEMODE_TTS_BASE_URLS", "http://127.0.0.1:8880/v1,https://api.openai.com/v1")
 STT_BASE_URLS = parse_comma_list("VOICEMODE_STT_BASE_URLS", "http://127.0.0.1:2022/v1,https://api.openai.com/v1")
-TTS_VOICES = parse_comma_list("VOICEMODE_TTS_VOICES", "af_sky,alloy")
+TTS_VOICES = parse_comma_list("VOICEMODE_VOICES", "af_sky,alloy")
 TTS_MODELS = parse_comma_list("VOICEMODE_TTS_MODELS", "tts-1,tts-1-hd,gpt-4o-mini-tts")
 
+# Voice preferences cache
+_cached_voice_preferences: Optional[list] = None
+_voice_preferences_loaded = False
+
+def get_voice_preferences() -> list[str]:
+    """
+    Get voice preferences from configuration.
+    
+    Uses the VOICEMODE_VOICES configuration which is loaded from:
+    1. Environment variables (highest priority)
+    2. Project-specific .voicemode.env files 
+    3. Global ~/.voicemode/voicemode.env file
+    4. Built-in defaults
+    
+    Returns:
+        List of voice names in preference order
+    """
+    global _cached_voice_preferences, _voice_preferences_loaded
+    
+    # Return cached preferences if already loaded
+    if _voice_preferences_loaded:
+        return _cached_voice_preferences or []
+    
+    _voice_preferences_loaded = True
+    
+    # Get voices from TTS_VOICES configuration
+    _cached_voice_preferences = TTS_VOICES.copy()
+    
+    logger.info(f"Voice preferences loaded: {_cached_voice_preferences}")
+    return _cached_voice_preferences
+
+def clear_voice_preferences_cache():
+    """Clear the voice preferences cache, forcing a reload on next access."""
+    global _cached_voice_preferences, _voice_preferences_loaded
+    _cached_voice_preferences = None
+    _voice_preferences_loaded = False
+    logger.debug("Voice preferences cache cleared")
+
+def reload_configuration():
+    """Reload configuration from files and clear all caches."""
+    # Clear voice preferences cache
+    clear_voice_preferences_cache()
+    
+    # Reload environment configuration
+    load_voicemode_env()
+    
+    # Update global configuration variables
+    global TTS_VOICES, TTS_MODELS, TTS_BASE_URLS, STT_BASE_URLS
+    TTS_BASE_URLS = parse_comma_list("VOICEMODE_TTS_BASE_URLS", "http://127.0.0.1:8880/v1,https://api.openai.com/v1")
+    STT_BASE_URLS = parse_comma_list("VOICEMODE_STT_BASE_URLS", "http://127.0.0.1:2022/v1,https://api.openai.com/v1")
+    TTS_VOICES = parse_comma_list("VOICEMODE_VOICES", "af_sky,alloy")
+    TTS_MODELS = parse_comma_list("VOICEMODE_TTS_MODELS", "tts-1,tts-1-hd,gpt-4o-mini-tts")
+    
+    logger.info("Configuration reloaded successfully")
+
 # Legacy variables have been removed - use the new list-based configuration:
 # - VOICEMODE_TTS_BASE_URLS (comma-separated list)
 # - VOICEMODE_STT_BASE_URLS (comma-separated list)
-# - VOICEMODE_TTS_VOICES (comma-separated list)
+# - VOICEMODE_VOICES (comma-separated list)
 # - VOICEMODE_TTS_MODELS (comma-separated list)
 
 # LiveKit configuration

{voice_mode-3.34.3 → voice_mode-4.0.1}/voice_mode/providers.py

@@ -9,9 +9,8 @@ import logging
 from typing import Dict, Optional, List, Any, Tuple
 from openai import AsyncOpenAI
 
-from .config import TTS_VOICES, TTS_MODELS, TTS_BASE_URLS, OPENAI_API_KEY
+from .config import TTS_VOICES, TTS_MODELS, TTS_BASE_URLS, OPENAI_API_KEY, get_voice_preferences
 from .provider_discovery import provider_registry, EndpointInfo
-from .voice_preferences import get_preferred_voices
 
 logger = logging.getLogger("voice-mode")
 
@@ -68,14 +67,14 @@ async def get_tts_client_and_voice(
         return client, selected_voice, selected_model, endpoint_info
     
     # Voice-first selection algorithm
-    # Get user preferences and prepend to system defaults
-    user_preferences = get_preferred_voices()
-    combined_voice_list = user_preferences + [v for v in TTS_VOICES if v not in user_preferences]
+    # Get user preferences from configuration
+    voice_preferences = get_voice_preferences()
+    combined_voice_list = voice_preferences
     
     logger.info(f"TTS Provider Selection (voice-first)")
-    if user_preferences:
-        logger.info(f"  User voice preferences: {user_preferences}")
-    logger.info(f"  Combined voice list: {combined_voice_list}")
+    if voice_preferences:
+        logger.info(f"  Voice preferences: {voice_preferences}")
+    logger.info(f"  Voice list: {combined_voice_list}")
     logger.info(f"  Preferred models: {TTS_MODELS}")
     logger.info(f"  Available endpoints: {TTS_BASE_URLS}")
     

{voice_mode-3.34.3 → voice_mode-4.0.1}/voice_mode/resources/configuration.py

@@ -267,7 +267,7 @@ async def environment_variables() -> str:
         ("VOICEMODE_AUTO_START_KOKORO", "Auto-start Kokoro service (true/false)"),
         ("VOICEMODE_TTS_BASE_URLS", "Comma-separated list of TTS endpoints"),
         ("VOICEMODE_STT_BASE_URLS", "Comma-separated list of STT endpoints"),
-        ("VOICEMODE_TTS_VOICES", "Comma-separated list of preferred voices"),
+        ("VOICEMODE_VOICES", "Comma-separated list of preferred voices"),
         ("VOICEMODE_TTS_MODELS", "Comma-separated list of preferred models"),
         # Audio Settings
         ("VOICEMODE_AUDIO_FORMAT", "Audio format for recording (pcm/mp3/wav/flac/aac/opus)"),
@@ -358,7 +358,7 @@ async def environment_template() -> str:
         f"export VOICEMODE_AUTO_START_KOKORO=\"{str(AUTO_START_KOKORO).lower()}\"",
         f"export VOICEMODE_TTS_BASE_URLS=\"{','.join(TTS_BASE_URLS)}\"",
         f"export VOICEMODE_STT_BASE_URLS=\"{','.join(STT_BASE_URLS)}\"",
-        f"export VOICEMODE_TTS_VOICES=\"{','.join(TTS_VOICES)}\"",
+        f"export VOICEMODE_VOICES=\"{','.join(TTS_VOICES)}\"",
         f"export VOICEMODE_TTS_MODELS=\"{','.join(TTS_MODELS)}\"",
         "",
         "# Audio Settings",