100 lines
3.0 KiB
YAML
100 lines
3.0 KiB
YAML
# Voice CLI Server Configuration with TTS support
|
|
|
|
server:
|
|
# Server binding address (0.0.0.0 for all interfaces)
|
|
host: "0.0.0.0"
|
|
# Server port
|
|
port: 8087
|
|
# Maximum file size for uploads (in bytes) - 200MB default
|
|
max_file_size: 209715200
|
|
# Enable CORS for web browser access
|
|
cors_enabled: true
|
|
|
|
whisper:
|
|
# Default model to use for transcription
|
|
default_model: "large-v3"
|
|
# Directory to store whisper models
|
|
models_dir: "./models"
|
|
# Automatically download models when needed
|
|
auto_download: true
|
|
# List of supported whisper models
|
|
supported_models:
|
|
- "tiny"
|
|
- "tiny.en"
|
|
- "base"
|
|
- "base.en"
|
|
- "small"
|
|
- "small.en"
|
|
- "medium"
|
|
- "medium.en"
|
|
- "large-v1"
|
|
- "large-v2"
|
|
- "large-v3"
|
|
|
|
# Audio processing settings
|
|
audio_processing:
|
|
supported_formats: ["mp3", "wav", "flac", "m4a", "ogg", "aac", "opus", "amr", "wma", "aiff", "caf", "mp4", "mov", "avi", "mkv", "webm", "3gp", "flv", "wmv", "mpeg", "mxf"]
|
|
auto_convert: true
|
|
conversion_timeout: 60 # seconds
|
|
temp_file_cleanup: true
|
|
temp_file_retention: 300 # 5 minutes
|
|
|
|
# Worker-based concurrency settings
|
|
workers:
|
|
transcription_workers: 3 # Number of worker threads for transcription
|
|
channel_buffer_size: 100 # Channel buffer size for task queue
|
|
worker_timeout: 3600 # Worker processing timeout in seconds
|
|
|
|
# TTS (Text-to-Speech) configuration
|
|
tts:
|
|
# Python interpreter path (relative to project directory)
|
|
python_path: ".venv/bin/python"
|
|
# TTS model directory (optional, for advanced TTS models)
|
|
model_path: "./checkpoints"
|
|
# Default voice model
|
|
default_model: "default"
|
|
# Supported audio formats for TTS output
|
|
supported_formats: ["mp3", "wav"]
|
|
# Maximum text length for TTS processing
|
|
max_text_length: 5000
|
|
# Default speech speed (0.5 to 2.0)
|
|
default_speed: 1.0
|
|
# Default pitch adjustment (-20 to 20)
|
|
default_pitch: 0
|
|
# Default volume (0.5 to 2.0)
|
|
default_volume: 1.0
|
|
# TTS processing timeout in seconds
|
|
timeout_seconds: 300
|
|
# Path to TTS service script (relative to project directory)
|
|
script_path: "tts_service.py"
|
|
|
|
logging:
|
|
# Log level (trace, debug, info, warn, error)
|
|
level: "info"
|
|
# Directory for log files
|
|
log_dir: "./logs"
|
|
# Maximum size per log file
|
|
max_file_size: "100MB"
|
|
# Maximum number of log files to keep
|
|
max_files: 20
|
|
|
|
daemon:
|
|
# PID file for daemon mode
|
|
pid_file: "./voice-cli-server.pid"
|
|
# Log file for daemon output
|
|
log_file: "./logs/server-daemon.log"
|
|
# Working directory for daemon
|
|
work_dir: "./"
|
|
|
|
# Async task management configuration
|
|
task_management:
|
|
|
|
# Apalis worker configuration
|
|
max_concurrent_tasks: 4 # Max concurrent async tasks
|
|
retry_attempts: 2 # Number of retry attempts for failed tasks
|
|
task_timeout_seconds: 3600 # Task processing timeout in seconds
|
|
catch_panic: true # Catch panics in execution and pipe them as errors
|
|
task_retention_minutes: 1440 # task retention in minutes (1440 minutes = 24 hours)
|
|
|
|
# SQLite database configuration
|
|
sqlite_db_path: "./data/tasks.db" # Path to SQLite database file |