A fundamental AIGC toolkit for music, song and audio generation using PyTorch
A comprehensive solution for music, song, and audio generation through advanced audio tokenization
Powerful control system lets you shape output with text prompts, music genres, and structures
Convenient scripts for model fine-tuning and inference to match your specific needs
Premium 24kHz mono output for text-to-music and music continuation tasks
Get started with Inspire Audio by installing via pip:
pip install inspire-audioAccess InspireAudio's powerful pre-trained models:
from inspire_audio import download_models
download_models('inspire-audio-base')Generate music with just a few lines of code:
from inspire_audio import AudioGenerator
generator = AudioGenerator('inspire-audio-base')
audio = generator.generate(
prompt="A peaceful piano melody",
duration=30 # seconds
)Save the generated audio to a file:
audio.save('output.wav')generator = AudioGenerator('inspire-audio-base')
audio = generator.generate(
prompt="An energetic rock song",
genre="rock",
bpm=120,
duration=60
)Inspire Audio offers a range of powerful models to suit different needs. The InspireAudio model family includes:
| Model Name | Description |
|---|---|
| InspireAudio-Base | Pre-trained Music Generation Model, 24kHz mono |
| InspireAudio-1.5B | Pre-trained Music Generation 1.5B Model, 24kHz mono |
| InspireSong-Base | Pre-trained Song Generation Base Model, 24kHz mono |
Inspire Audio is a fundamental AIGC toolkit built on PyTorch that enables music, song, and audio generation through advanced audio tokenization techniques.
Currently, InspireAudio supports music generation and song generation with 24kHz mono output. It can generate audio based on text prompts and continue existing music.
You can control the output using text prompts, specify music genres, and define music structures to get your desired results.
We offer InspireAudio-Base, InspireAudio-1.5B for music generation, and InspireSong-Base for song generation. All models output 24kHz mono audio.
Yes, InspireAudio provides convenient scripts for model fine-tuning and inference to adapt the models to your specific needs.
You can contribute by starring our GitHub repository, reporting issues, submitting pull requests, or joining our community discussions.
InspireAudio combines state-of-the-art AI technology with an intuitive interface, making advanced audio generation accessible to everyone.