anyrobo 0.1.0__tar.gz

anyrobo-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Viet-Anh Nguyen
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE. 

anyrobo-0.1.0/PKG-INFO

@@ -0,0 +1,179 @@
+Metadata-Version: 2.4
+Name: anyrobo
+Version: 0.1.0
+Summary: Create your own JARVIS or GLADOS: a framework for voice-powered AI assistants with unique personalities
+Author-email: Viet-Anh Nguyen <vietanh.dev@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/vietanhdev/anyrobo
+Project-URL: Bug Tracker, https://github.com/vietanhdev/anyrobo/issues
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: MacOS
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy<3.0.0,>=2.0.0
+Requires-Dist: onnxruntime<2.0.0,>=1.16.0
+Requires-Dist: kokoro-onnx>=0.2.0
+Requires-Dist: lightning-whisper-mlx<0.1.0,>=0.0.10
+Requires-Dist: sounddevice<0.6.0,>=0.4.6
+Requires-Dist: ollama<0.5.0,>=0.4.0
+Dynamic: license-file
+
+# AnyRobo - Create Your Own Robo Assistant
+
+AnyRobo is an advanced speech-to-speech AI assistant framework that enables you to create your own real-life version of sci-fi AI assistants like JARVIS (from Iron Man) or GLADOS (from Portal). Powered by state-of-the-art machine learning models, AnyRobo listens to your voice, understands your requests, and responds with natural-sounding speech in real-time.
+
+![AnyRobo](https://img.shields.io/badge/AnyRobo-0.1.0-blue)
+![Python](https://img.shields.io/badge/Python-3.10+-green)
+![License](https://img.shields.io/badge/License-MIT-yellow)
+
+## Why AnyRobo?
+
+Have you ever wanted to create your own JARVIS or GLADOS? AnyRobo provides a fully modular framework that allows you to:
+
+- Create voice-powered AI assistants with unique personalities
+- Customize voice profiles to sound like your favorite AI characters
+- Build advanced conversation capabilities with state-of-the-art language models
+- Deploy your assistant on macOS with optimized performance for Apple Silicon
+
+## Core Technologies
+
+- **Speech Recognition**: [Whisper MLX](https://github.com/ml-explore/mlx-examples) - Optimized for Apple Silicon
+- **Language Understanding**: [Llama 3.2](https://ollama.com/library/llama3.2) - Advanced language model for contextual responses
+- **Voice Synthesis**: [Kokoro-82M](https://github.com/thewh1teagle/kokoro-onnx) - High-quality text-to-speech
+
+## Features
+
+- **Continuous Listening**: Automatically detects when you've finished speaking
+- **Natural Conversations**: Responds intelligently to a wide range of queries and commands
+- **Real-time Synthesis**: Generates human-like speech with minimal latency
+- **Voice Customization**: Supports multiple voice profiles
+- **Streaming Responses**: Begins speaking before the full response is generated
+- **Optimized Performance**: Designed for efficiency on Apple Silicon
+
+## Installation
+
+### Quick Install (from PyPI)
+
+```bash
+pip install anyrobo
+```
+
+### Install from Source
+
+```bash
+git clone https://github.com/vietanhdev/anyrobo.git
+cd anyrobo
+pip install -e .
+```
+
+### Setup Dependencies
+
+AnyRobo requires [Ollama](https://ollama.com/) for LLM support:
+
+```bash
+# Install Ollama
+curl -fsSL https://ollama.com/install.sh | sh
+# Pull the required model
+ollama pull llama3.2
+```
+
+## Usage
+
+### Command-line Interface
+
+```bash
+# Download required models and start the assistant
+anyrobo --setup
+anyrobo
+```
+
+With custom settings:
+
+```bash
+anyrobo --voice am_michael --speed 1.3 --silence-threshold 0.03
+```
+
+### As a Library
+
+```python
+from anyrobo import AnyRobo
+from anyrobo.models.loader import download_tts_model, ensure_ollama_model
+
+# Download required models
+download_tts_model()
+ensure_ollama_model("llama3.2")
+
+# Create and run assistant
+assistant = AnyRobo(
+    voice="am_michael",
+    speed=1.2,
+    system_prompt=(
+        "You are J.A.R.V.I.S., an advanced AI assistant. "
+        "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+    )
+)
+
+# Start listening and responding
+assistant.record_and_transcribe()
+```
+
+## Create Your Own AI Character
+
+You can customize the personality of your assistant by modifying the system prompt:
+
+```python
+# JARVIS from Iron Man
+system_prompt = (
+    "You are J.A.R.V.I.S., an advanced AI assistant. "
+    "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+)
+
+# GLADOS from Portal
+system_prompt = (
+    "You are GLaDOS, an AI with a dark sense of humor. "
+    "Respond to queries sarcastically, occasionally mentioning cake or testing."
+)
+
+# HAL 9000 from 2001: A Space Odyssey
+system_prompt = (
+    "You are HAL 9000. Be calm, logical, and slightly ominous in your responses. "
+    "Speak in a slow, deliberate manner and be excessively literal."
+)
+```
+
+## Configuration Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `voice` | Voice profile to use | `"am_michael"` |
+| `speed` | Speed factor for speech | `1.2` |
+| `silence_threshold` | Volume level that counts as silence | `0.02` |
+| `silence_duration` | Seconds of silence before cutting recording | `1.5` |
+| `sample_rate` | Audio sample rate in Hz | `24000` |
+| `system_prompt` | Custom system prompt for the LLM | *See code* |
+
+## Troubleshooting
+
+- **No audio output**: Ensure your system audio output is correctly configured
+- **Poor recognition**: Try speaking more clearly or adjust the `silence_threshold` value
+- **Model loading issues**: Run `anyrobo --setup` to download all required models
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.
+
+## Contributing
+
+Contributions are welcome! Please feel free to submit a Pull Request.
+
+## Acknowledgements
+
+AnyRobo is built on top of several open-source projects and pre-trained models. We're grateful to the developers and researchers who make their work available to the community.
+

anyrobo-0.1.0/README.md

@@ -0,0 +1,152 @@
+# AnyRobo - Create Your Own Robo Assistant
+
+AnyRobo is an advanced speech-to-speech AI assistant framework that enables you to create your own real-life version of sci-fi AI assistants like JARVIS (from Iron Man) or GLADOS (from Portal). Powered by state-of-the-art machine learning models, AnyRobo listens to your voice, understands your requests, and responds with natural-sounding speech in real-time.
+
+![AnyRobo](https://img.shields.io/badge/AnyRobo-0.1.0-blue)
+![Python](https://img.shields.io/badge/Python-3.10+-green)
+![License](https://img.shields.io/badge/License-MIT-yellow)
+
+## Why AnyRobo?
+
+Have you ever wanted to create your own JARVIS or GLADOS? AnyRobo provides a fully modular framework that allows you to:
+
+- Create voice-powered AI assistants with unique personalities
+- Customize voice profiles to sound like your favorite AI characters
+- Build advanced conversation capabilities with state-of-the-art language models
+- Deploy your assistant on macOS with optimized performance for Apple Silicon
+
+## Core Technologies
+
+- **Speech Recognition**: [Whisper MLX](https://github.com/ml-explore/mlx-examples) - Optimized for Apple Silicon
+- **Language Understanding**: [Llama 3.2](https://ollama.com/library/llama3.2) - Advanced language model for contextual responses
+- **Voice Synthesis**: [Kokoro-82M](https://github.com/thewh1teagle/kokoro-onnx) - High-quality text-to-speech
+
+## Features
+
+- **Continuous Listening**: Automatically detects when you've finished speaking
+- **Natural Conversations**: Responds intelligently to a wide range of queries and commands
+- **Real-time Synthesis**: Generates human-like speech with minimal latency
+- **Voice Customization**: Supports multiple voice profiles
+- **Streaming Responses**: Begins speaking before the full response is generated
+- **Optimized Performance**: Designed for efficiency on Apple Silicon
+
+## Installation
+
+### Quick Install (from PyPI)
+
+```bash
+pip install anyrobo
+```
+
+### Install from Source
+
+```bash
+git clone https://github.com/vietanhdev/anyrobo.git
+cd anyrobo
+pip install -e .
+```
+
+### Setup Dependencies
+
+AnyRobo requires [Ollama](https://ollama.com/) for LLM support:
+
+```bash
+# Install Ollama
+curl -fsSL https://ollama.com/install.sh | sh
+# Pull the required model
+ollama pull llama3.2
+```
+
+## Usage
+
+### Command-line Interface
+
+```bash
+# Download required models and start the assistant
+anyrobo --setup
+anyrobo
+```
+
+With custom settings:
+
+```bash
+anyrobo --voice am_michael --speed 1.3 --silence-threshold 0.03
+```
+
+### As a Library
+
+```python
+from anyrobo import AnyRobo
+from anyrobo.models.loader import download_tts_model, ensure_ollama_model
+
+# Download required models
+download_tts_model()
+ensure_ollama_model("llama3.2")
+
+# Create and run assistant
+assistant = AnyRobo(
+    voice="am_michael",
+    speed=1.2,
+    system_prompt=(
+        "You are J.A.R.V.I.S., an advanced AI assistant. "
+        "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+    )
+)
+
+# Start listening and responding
+assistant.record_and_transcribe()
+```
+
+## Create Your Own AI Character
+
+You can customize the personality of your assistant by modifying the system prompt:
+
+```python
+# JARVIS from Iron Man
+system_prompt = (
+    "You are J.A.R.V.I.S., an advanced AI assistant. "
+    "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+)
+
+# GLADOS from Portal
+system_prompt = (
+    "You are GLaDOS, an AI with a dark sense of humor. "
+    "Respond to queries sarcastically, occasionally mentioning cake or testing."
+)
+
+# HAL 9000 from 2001: A Space Odyssey
+system_prompt = (
+    "You are HAL 9000. Be calm, logical, and slightly ominous in your responses. "
+    "Speak in a slow, deliberate manner and be excessively literal."
+)
+```
+
+## Configuration Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `voice` | Voice profile to use | `"am_michael"` |
+| `speed` | Speed factor for speech | `1.2` |
+| `silence_threshold` | Volume level that counts as silence | `0.02` |
+| `silence_duration` | Seconds of silence before cutting recording | `1.5` |
+| `sample_rate` | Audio sample rate in Hz | `24000` |
+| `system_prompt` | Custom system prompt for the LLM | *See code* |
+
+## Troubleshooting
+
+- **No audio output**: Ensure your system audio output is correctly configured
+- **Poor recognition**: Try speaking more clearly or adjust the `silence_threshold` value
+- **Model loading issues**: Run `anyrobo --setup` to download all required models
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.
+
+## Contributing
+
+Contributions are welcome! Please feel free to submit a Pull Request.
+
+## Acknowledgements
+
+AnyRobo is built on top of several open-source projects and pre-trained models. We're grateful to the developers and researchers who make their work available to the community.
+

anyrobo-0.1.0/anyrobo/__init__.py

@@ -0,0 +1,7 @@
+"""AnyRobo - Your Voice-Powered AI Assistant"""
+
+__version__ = "0.1.0"
+
+from anyrobo.assistant import AnyRobo
+
+__all__ = ["AnyRobo"]

anyrobo-0.1.0/anyrobo/assistant.py

@@ -0,0 +1,228 @@
+"""Core assistant module for AnyRobo."""
+
+import signal
+from concurrent.futures import ThreadPoolExecutor
+from threading import Event
+from typing import Dict, List, Optional
+
+import sounddevice as sd
+import numpy as np
+from ollama import chat
+
+from anyrobo.speech.recognition import SpeechRecognizer
+from anyrobo.speech.synthesis import TextToSpeech
+
+
+class AnyRobo:
+    """Main assistant class that coordinates speech recognition and synthesis."""
+    
+    def __init__(
+        self,
+        sample_rate: int = 24000,
+        silence_threshold: float = 0.02,
+        silence_duration: float = 1.5,
+        voice: str = "am_michael",
+        speed: float = 1.2,
+        system_prompt: Optional[str] = None,
+    ):
+        """Initialize the AnyRobo assistant.
+        
+        Args:
+            sample_rate: Audio sample rate in Hz
+            silence_threshold: Volume level that counts as silence
+            silence_duration: Seconds of silence before cutting recording
+            voice: Voice profile to use
+            speed: Speech speed factor
+            system_prompt: Custom system prompt for the LLM
+        """
+        # audio settings
+        self.SAMPLE_RATE = sample_rate
+        self.SILENCE_THRESHOLD = silence_threshold
+        self.SILENCE_DURATION = silence_duration
+
+        # text-to-speech settings
+        self.SPEED = speed
+        self.VOICE = voice
+        self.CHUNK_SIZE = 300  # size of text chunks for processing
+
+        # ollama settings
+        self.messages = []
+        self.SYSTEM_PROMPT = system_prompt or "Give a conversational response to the following statement or question in 1-2 sentences. The response should be natural and engaging, and the length depends on what you have to say."
+
+        # init components
+        self.speech_recognizer = SpeechRecognizer(model="small", batch_size=12)
+        self.tts = TextToSpeech()
+        self.executor = ThreadPoolExecutor(max_workers=1)
+
+        # interrupt handling
+        self.shutdown_event = Event()
+        signal.signal(signal.SIGINT, self._signal_handler)
+
+    def _signal_handler(self, signum, frame):
+        """Handle interrupt signals."""
+        print("\nStopping...")
+        self.shutdown_event.set()
+
+    def record_and_transcribe(self):
+        """Main loop: record audio, transcribe, and respond."""
+        # state for audio recording
+        audio_buffer = []
+        silence_frames = 0
+        total_frames = 0
+
+        def callback(indata, frames, time_info, status):
+            # callback function that processes incoming audio frames
+            if self.shutdown_event.is_set():
+                raise sd.CallbackStop()
+
+            nonlocal audio_buffer, silence_frames, total_frames
+
+            if status:
+                print(status)
+
+            audio = indata.flatten()
+            level = np.abs(audio).mean()
+
+            audio_buffer.extend(audio.tolist())
+            total_frames += len(audio)
+
+            # track silence duration
+            if level < self.SILENCE_THRESHOLD:
+                silence_frames += len(audio)
+            else:
+                silence_frames = 0
+
+            # process audio when silence is detected
+            if silence_frames > self.SILENCE_DURATION * self.SAMPLE_RATE:
+                audio_segment = np.array(audio_buffer, dtype=np.float32)
+
+                if len(audio_segment) > self.SAMPLE_RATE:
+                    text = self.speech_recognizer.transcribe(audio_segment)['text']
+
+                    # skip empty/invalid transcriptions
+                    if text.strip():
+                        print(f"Transcription: {text}")
+                        self.messages.append({
+                            'role': 'user',
+                            'content': text
+                        })
+                        self.create_and_play_response(text)
+
+                # reset state
+                audio_buffer.clear()
+                silence_frames = 0
+                total_frames = 0
+
+        # start recording loop
+        try:
+            with sd.InputStream(
+                callback=callback,
+                channels=1,
+                samplerate=self.SAMPLE_RATE,
+                dtype=np.float32
+            ):
+                print("Recording... Press Ctrl+C to stop")
+                while not self.shutdown_event.is_set():
+                    sd.sleep(100)
+        except sd.CallbackStop:
+            pass
+
+    def create_and_play_response(self, prompt: str):
+        """Generate and speak a response to the user's input."""
+        if self.shutdown_event.is_set():
+            return
+
+        # stream response from llm
+        stream = chat(
+            model='llama3.2',
+            messages=[{
+                'role': 'system',
+                'content': self.SYSTEM_PROMPT
+            }] + self.messages,
+            stream=True,
+        )
+
+        # state for processing response
+        futures = []
+        buffer = ""
+        curr_str = ""
+
+        try:
+            # process response stream
+            for chunk in stream:
+                if self.shutdown_event.is_set():
+                    break
+
+                print(chunk)
+                text = chunk['message']['content']
+
+                if len(text) == 0:
+                    self.messages.append({
+                        'role': 'assistant',
+                        'content': curr_str
+                    })
+                    curr_str = ""
+                    print(self.messages)
+                    continue
+
+                buffer += text
+                curr_str += text
+
+                # find end of sentence to chunk at
+                last_punctuation = max(
+                    buffer.rfind('. '),
+                    buffer.rfind('? '),
+                    buffer.rfind('! ')
+                )
+
+                if last_punctuation == -1:
+                    continue
+
+                # handle long chunks
+                while last_punctuation != -1 and last_punctuation >= self.CHUNK_SIZE:
+                    last_punctuation = max(
+                        buffer.rfind(', ', 0, last_punctuation),
+                        buffer.rfind('; ', 0, last_punctuation),
+                        buffer.rfind('— ', 0, last_punctuation)
+                    )
+
+                if last_punctuation == -1:
+                    last_punctuation = buffer.find(' ', 0, self.CHUNK_SIZE)
+
+                # process chunk
+                # convert chunk to audio
+                chunk_text = buffer[:last_punctuation + 1]
+                futures.append(
+                    self.executor.submit(
+                        self.tts.generate_audio,
+                        chunk_text, self.VOICE, self.SPEED
+                    )
+                )
+                buffer = buffer[last_punctuation + 1:]
+
+            # process final chunk if any
+            if buffer and not self.shutdown_event.is_set():
+                futures.append(
+                    self.executor.submit(
+                        self.tts.generate_audio,
+                        buffer, self.VOICE, self.SPEED
+                    )
+                )
+
+            # play generated audio
+            if not self.shutdown_event.is_set():
+                with sd.OutputStream(
+                    samplerate=self.SAMPLE_RATE,
+                    channels=1,
+                    dtype=np.float32
+                ) as out_stream:
+                    for fut in futures:
+                        if self.shutdown_event.is_set():
+                            break
+                        audio_data = fut.result()
+                        if len(audio_data) == 0:
+                            continue
+                        out_stream.write(audio_data.reshape(-1, 1))
+        except Exception as e:
+            if not self.shutdown_event.is_set():
+                raise e 

anyrobo-0.1.0/anyrobo/cli.py

@@ -0,0 +1,146 @@
+"""Command-line interface for AnyRobo."""
+
+import argparse
+import os
+import sys
+from typing import List, Optional
+
+from anyrobo import AnyRobo
+from anyrobo.models.loader import (
+    download_tts_model,
+    download_whisper_model,
+    ensure_ollama_model
+)
+
+
+def parse_args(args: Optional[List[str]] = None) -> argparse.Namespace:
+    """Parse command-line arguments.
+    
+    Args:
+        args: Command-line arguments
+        
+    Returns:
+        Parsed arguments
+    """
+    parser = argparse.ArgumentParser(
+        description="AnyRobo - Voice-powered AI assistant"
+    )
+    
+    parser.add_argument(
+        "--voice", 
+        type=str, 
+        default="af_sarah",
+        help="Voice profile to use (default: af_sarah)"
+    )
+    
+    parser.add_argument(
+        "--speed", 
+        type=float, 
+        default=1.2,
+        help="Speed factor for speech (default: 1.2)"
+    )
+    
+    parser.add_argument(
+        "--silence-threshold", 
+        type=float, 
+        default=0.02,
+        help="Volume level that counts as silence (default: 0.02)"
+    )
+    
+    parser.add_argument(
+        "--silence-duration", 
+        type=float, 
+        default=1.5,
+        help="Seconds of silence before cutting recording (default: 1.5)"
+    )
+    
+    parser.add_argument(
+        "--sample-rate", 
+        type=int, 
+        default=24000,
+        help="Audio sample rate in Hz (default: 24000)"
+    )
+    
+    parser.add_argument(
+        "--model", 
+        type=str, 
+        default="llama3.2",
+        help="Ollama model to use (default: llama3.2)"
+    )
+    
+    parser.add_argument(
+        "--prompt", 
+        type=str, 
+        default=None,
+        help="Custom system prompt for the LLM"
+    )
+    
+    parser.add_argument(
+        "--setup", 
+        action="store_true",
+        help="Download required models without starting the assistant"
+    )
+    
+    return parser.parse_args(args)
+
+
+def setup_models() -> None:
+    """Download required models."""
+    print("Setting up AnyRobo...")
+    tts_model_path = download_tts_model()
+    whisper_model_dir = download_whisper_model("small")
+    ensure_ollama_model("llama3.2")
+    print("\nSetup complete! You can now run AnyRobo.")
+
+
+def main(args: Optional[List[str]] = None) -> int:
+    """Main entry point for AnyRobo.
+    
+    Args:
+        args: Command-line arguments
+        
+    Returns:
+        Exit code
+    """
+    parsed_args = parse_args(args)
+    
+    # Just setup models if requested
+    if parsed_args.setup:
+        setup_models()
+        return 0
+    
+    # Create and run the assistant
+    try:
+        # Make sure models are available
+        tts_model_path = download_tts_model()
+        whisper_model_dir = download_whisper_model("small")
+        ensure_ollama_model(parsed_args.model)
+        
+        # Create the assistant
+        assistant = AnyRobo(
+            sample_rate=parsed_args.sample_rate,
+            silence_threshold=parsed_args.silence_threshold,
+            silence_duration=parsed_args.silence_duration,
+            voice=parsed_args.voice,
+            speed=parsed_args.speed,
+            system_prompt=parsed_args.prompt
+        )
+        
+        print(f"Starting AnyRobo - your voice-powered AI assistant...")
+        print(f"Using voice: {parsed_args.voice}")
+        print(f"Using model: {parsed_args.model}")
+        print("Press Ctrl+C to stop")
+        
+        # Run the assistant
+        assistant.record_and_transcribe()
+        return 0
+    except KeyboardInterrupt:
+        print("\nStopping AnyRobo...")
+        return 0
+    except Exception as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+
+if __name__ == "__main__":
+    sys.exit(main()) 

anyrobo-0.1.0/anyrobo.egg-info/PKG-INFO

@@ -0,0 +1,179 @@
+Metadata-Version: 2.4
+Name: anyrobo
+Version: 0.1.0
+Summary: Create your own JARVIS or GLADOS: a framework for voice-powered AI assistants with unique personalities
+Author-email: Viet-Anh Nguyen <vietanh.dev@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/vietanhdev/anyrobo
+Project-URL: Bug Tracker, https://github.com/vietanhdev/anyrobo/issues
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: MacOS
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy<3.0.0,>=2.0.0
+Requires-Dist: onnxruntime<2.0.0,>=1.16.0
+Requires-Dist: kokoro-onnx>=0.2.0
+Requires-Dist: lightning-whisper-mlx<0.1.0,>=0.0.10
+Requires-Dist: sounddevice<0.6.0,>=0.4.6
+Requires-Dist: ollama<0.5.0,>=0.4.0
+Dynamic: license-file
+
+# AnyRobo - Create Your Own Robo Assistant
+
+AnyRobo is an advanced speech-to-speech AI assistant framework that enables you to create your own real-life version of sci-fi AI assistants like JARVIS (from Iron Man) or GLADOS (from Portal). Powered by state-of-the-art machine learning models, AnyRobo listens to your voice, understands your requests, and responds with natural-sounding speech in real-time.
+
+![AnyRobo](https://img.shields.io/badge/AnyRobo-0.1.0-blue)
+![Python](https://img.shields.io/badge/Python-3.10+-green)
+![License](https://img.shields.io/badge/License-MIT-yellow)
+
+## Why AnyRobo?
+
+Have you ever wanted to create your own JARVIS or GLADOS? AnyRobo provides a fully modular framework that allows you to:
+
+- Create voice-powered AI assistants with unique personalities
+- Customize voice profiles to sound like your favorite AI characters
+- Build advanced conversation capabilities with state-of-the-art language models
+- Deploy your assistant on macOS with optimized performance for Apple Silicon
+
+## Core Technologies
+
+- **Speech Recognition**: [Whisper MLX](https://github.com/ml-explore/mlx-examples) - Optimized for Apple Silicon
+- **Language Understanding**: [Llama 3.2](https://ollama.com/library/llama3.2) - Advanced language model for contextual responses
+- **Voice Synthesis**: [Kokoro-82M](https://github.com/thewh1teagle/kokoro-onnx) - High-quality text-to-speech
+
+## Features
+
+- **Continuous Listening**: Automatically detects when you've finished speaking
+- **Natural Conversations**: Responds intelligently to a wide range of queries and commands
+- **Real-time Synthesis**: Generates human-like speech with minimal latency
+- **Voice Customization**: Supports multiple voice profiles
+- **Streaming Responses**: Begins speaking before the full response is generated
+- **Optimized Performance**: Designed for efficiency on Apple Silicon
+
+## Installation
+
+### Quick Install (from PyPI)
+
+```bash
+pip install anyrobo
+```
+
+### Install from Source
+
+```bash
+git clone https://github.com/vietanhdev/anyrobo.git
+cd anyrobo
+pip install -e .
+```
+
+### Setup Dependencies
+
+AnyRobo requires [Ollama](https://ollama.com/) for LLM support:
+
+```bash
+# Install Ollama
+curl -fsSL https://ollama.com/install.sh | sh
+# Pull the required model
+ollama pull llama3.2
+```
+
+## Usage
+
+### Command-line Interface
+
+```bash
+# Download required models and start the assistant
+anyrobo --setup
+anyrobo
+```
+
+With custom settings:
+
+```bash
+anyrobo --voice am_michael --speed 1.3 --silence-threshold 0.03
+```
+
+### As a Library
+
+```python
+from anyrobo import AnyRobo
+from anyrobo.models.loader import download_tts_model, ensure_ollama_model
+
+# Download required models
+download_tts_model()
+ensure_ollama_model("llama3.2")
+
+# Create and run assistant
+assistant = AnyRobo(
+    voice="am_michael",
+    speed=1.2,
+    system_prompt=(
+        "You are J.A.R.V.I.S., an advanced AI assistant. "
+        "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+    )
+)
+
+# Start listening and responding
+assistant.record_and_transcribe()
+```
+
+## Create Your Own AI Character
+
+You can customize the personality of your assistant by modifying the system prompt:
+
+```python
+# JARVIS from Iron Man
+system_prompt = (
+    "You are J.A.R.V.I.S., an advanced AI assistant. "
+    "Respond with a mix of helpfulness, light sarcasm, and technical prowess."
+)
+
+# GLADOS from Portal
+system_prompt = (
+    "You are GLaDOS, an AI with a dark sense of humor. "
+    "Respond to queries sarcastically, occasionally mentioning cake or testing."
+)
+
+# HAL 9000 from 2001: A Space Odyssey
+system_prompt = (
+    "You are HAL 9000. Be calm, logical, and slightly ominous in your responses. "
+    "Speak in a slow, deliberate manner and be excessively literal."
+)
+```
+
+## Configuration Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `voice` | Voice profile to use | `"am_michael"` |
+| `speed` | Speed factor for speech | `1.2` |
+| `silence_threshold` | Volume level that counts as silence | `0.02` |
+| `silence_duration` | Seconds of silence before cutting recording | `1.5` |
+| `sample_rate` | Audio sample rate in Hz | `24000` |
+| `system_prompt` | Custom system prompt for the LLM | *See code* |
+
+## Troubleshooting
+
+- **No audio output**: Ensure your system audio output is correctly configured
+- **Poor recognition**: Try speaking more clearly or adjust the `silence_threshold` value
+- **Model loading issues**: Run `anyrobo --setup` to download all required models
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details.
+
+## Contributing
+
+Contributions are welcome! Please feel free to submit a Pull Request.
+
+## Acknowledgements
+
+AnyRobo is built on top of several open-source projects and pre-trained models. We're grateful to the developers and researchers who make their work available to the community.
+

anyrobo-0.1.0/anyrobo.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+LICENSE
+README.md
+pyproject.toml
+setup.py
+anyrobo/__init__.py
+anyrobo/assistant.py
+anyrobo/cli.py
+anyrobo.egg-info/PKG-INFO
+anyrobo.egg-info/SOURCES.txt
+anyrobo.egg-info/dependency_links.txt
+anyrobo.egg-info/entry_points.txt
+anyrobo.egg-info/requires.txt
+anyrobo.egg-info/top_level.txt
+tests/test_assistant.py

anyrobo-0.1.0/anyrobo.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

anyrobo-0.1.0/anyrobo.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+anyrobo = anyrobo.cli:main

anyrobo-0.1.0/anyrobo.egg-info/requires.txt

@@ -0,0 +1,6 @@
+numpy<3.0.0,>=2.0.0
+onnxruntime<2.0.0,>=1.16.0
+kokoro-onnx>=0.2.0
+lightning-whisper-mlx<0.1.0,>=0.0.10
+sounddevice<0.6.0,>=0.4.6
+ollama<0.5.0,>=0.4.0

anyrobo-0.1.0/anyrobo.egg-info/top_level.txt

@@ -0,0 +1 @@
+anyrobo

anyrobo-0.1.0/pyproject.toml

@@ -0,0 +1,56 @@
+[build-system]
+requires = ["setuptools>=42", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "anyrobo"
+version = "0.1.0"
+description = "Create your own JARVIS or GLADOS: a framework for voice-powered AI assistants with unique personalities"
+readme = "README.md"
+requires-python = ">=3.10"
+license = {text = "MIT"}
+authors = [
+    {name = "Viet-Anh Nguyen", email = "vietanh.dev@gmail.com"}
+]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: MacOS",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+    "Topic :: Multimedia :: Sound/Audio :: Speech"
+]
+dependencies = [
+    "numpy>=2.0.0,<3.0.0",
+    "onnxruntime>=1.16.0,<2.0.0",
+    "kokoro-onnx>=0.2.0",
+    "lightning-whisper-mlx>=0.0.10,<0.1.0",
+    "sounddevice>=0.4.6,<0.6.0",
+    "ollama>=0.4.0,<0.5.0"
+]
+
+[project.urls]
+"Homepage" = "https://github.com/vietanhdev/anyrobo"
+"Bug Tracker" = "https://github.com/vietanhdev/anyrobo/issues"
+
+[project.scripts]
+anyrobo = "anyrobo.cli:main"
+
+[tool.setuptools]
+packages = ["anyrobo"]
+
+[tool.black]
+line-length = 100
+
+[tool.isort]
+profile = "black"
+line_length = 100
+
+[tool.mypy]
+python_version = "3.10"
+warn_return_any = true
+warn_unused_configs = true
+disallow_untyped_defs = true
+disallow_incomplete_defs = true 

anyrobo-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

anyrobo-0.1.0/setup.py

@@ -0,0 +1,6 @@
+"""Simple setup.py for AnyRobo."""
+
+from setuptools import setup
+
+if __name__ == "__main__":
+    setup() 

anyrobo-0.1.0/tests/test_assistant.py

@@ -0,0 +1,53 @@
+"""Tests for AnyRobo."""
+
+import unittest
+from unittest.mock import MagicMock, patch
+
+import numpy as np
+
+from anyrobo import AnyRobo
+
+
+class TestAnyRobo(unittest.TestCase):
+    """Tests for the AnyRobo class."""
+
+    @patch('anyrobo.speech.recognition.SpeechRecognizer')
+    @patch('anyrobo.speech.synthesis.TextToSpeech')
+    def test_init(self, mock_tts, mock_recognizer):
+        """Test initialization with default parameters."""
+        assistant = AnyRobo()
+        
+        self.assertEqual(assistant.SAMPLE_RATE, 24000)
+        self.assertEqual(assistant.SILENCE_THRESHOLD, 0.02)
+        self.assertEqual(assistant.SILENCE_DURATION, 1.5)
+        self.assertEqual(assistant.VOICE, "am_michael")
+        self.assertEqual(assistant.SPEED, 1.2)
+        self.assertEqual(assistant.CHUNK_SIZE, 300)
+        
+        # Ensure speech components were initialized
+        self.assertTrue(mock_recognizer.called)
+        self.assertTrue(mock_tts.called)
+        
+    @patch('anyrobo.speech.recognition.SpeechRecognizer')
+    @patch('anyrobo.speech.synthesis.TextToSpeech')
+    def test_custom_init(self, mock_tts, mock_recognizer):
+        """Test initialization with custom parameters."""
+        assistant = AnyRobo(
+            sample_rate=44100,
+            silence_threshold=0.05,
+            silence_duration=2.0,
+            voice="custom_voice",
+            speed=1.5,
+            system_prompt="Custom prompt",
+        )
+        
+        self.assertEqual(assistant.SAMPLE_RATE, 44100)
+        self.assertEqual(assistant.SILENCE_THRESHOLD, 0.05)
+        self.assertEqual(assistant.SILENCE_DURATION, 2.0)
+        self.assertEqual(assistant.VOICE, "custom_voice")
+        self.assertEqual(assistant.SPEED, 1.5)
+        self.assertEqual(assistant.SYSTEM_PROMPT, "Custom prompt")
+        
+
+if __name__ == "__main__":
+    unittest.main()