AL1S

Hehe, Alice is so cute. Anyway, Alice is on duty at Schale today~ 🌸

English | 中文简体

---

✨ Features

• 🎤 Voice wake-up ("爱丽丝")
• 🗣️ Multilingual speech recognition (if using supported AI model)
• 🤖 AI-based chat capabilities (OpenAI API)
• 🔊 GPT-Sovits text-to-speech
• 🔍 MQTT control, web search, etc., based on Function Calling API

🛠️ Requirements

• 🐍 Python 3.10+
• 🎤 Microphone
• (Required) uv Python package manager
• (Required) VLC Media Player
• (Required) OpenAI API key
• (Required) GPT-Sovits API
• (Required) Picovoice Access Key
• (Optional) Speech-To-Text AI API (e.g., SenseVoice)
• (Optional) MQTT server

📦 Installation

1. Clone this repository:

   git clone --depth 1 https://github.com/Archive202/AL1S.git
   cd AL1S

2. Install the vlc runtime librar (dependency for python-vlc):

   • Windows: Download from offical website
   • Linux (apt): sudo apt install vlc
   • macOS (brew): brew install vlc

3. Install the portaudio runtime library (dependency for pyaudio):

   • Linux (apt): sudo apt install portaudio19-dev
   • macOS: brew install portaudio

4. Use uv to install dependencies:

   uv sync

⚙️ Configuration

1. Copy the configuration template .env.example to .env:

   cp .env.example .env

2. Configure the .env file:

   # OpenAI API configuration
   OPENAI_API_KEY="YOUR_OPENAI_API_KEY"
   OPENAI_BASE_URL="https://api.openai.com/v1"
   OPENAI_MODEL="gpt-4o-mini"
   
   # Speech recognition configuration
   ## Wake-up
   PLATFORM="YOUR_PLATFORM"   # [windows, mac, linux]
   PICOVOICE_ACCESS_KEY="YOUR_PICOVOICE_ACCESS_KEY"   # Obtain from Picovoice website
   ## Speech recognition
   RECOGNIZER_ENGINE="google"                         # [google, custom]
   CUSTOM_RECOGNIZER_API_KEY="YOUR_CUSTOM_RECOGNIZER_API_KEY"  # For custom
   CUSTOM_RECOGNIZER_BASE_URL="CUSTOM_RECOGNIZER_BASE_URL"     # For custom
   CUSTOM_RECOGNIZER_MODEL="YOUR_CUSTOM_RECOGNIZER_MODEL"      # For custom           
   
   # MQTT configuration
   MQTT_BROKER="example.com"
   MQTT_PORT=1883
   MQTT_USERNAME="YOUR_MQTT_USERNAME"
   MQTT_PASSWORD="YOUR_MQTT_PASSWORD"
   
   # GSV configuration
   GSV_BASE_URL="YOUR_GSV_BASE_URL"            # GPT-Sovits API server address
   REF_AUDIO_PATH="YOUR_REF_AUDIO_PATH"        # Reference audio path (on the server)
   PROMPT_TEXT="YOUR_PROMPT_TEXT"              # Reference audio text
   PROMPT_LANG="YOUR_PROMPT_LANG"              # Reference audio language

🚀 Usage Guide

1. Alice is online!:

   uv run main.py

2. After starting, say the wake-up word 爱丽丝 to start chatting

3. Speak commands or start chatting

4. End the conversation by saying goodbye in the form of bye-bye, goodbye, etc

📝 License

MIT License - LICENSE