Instructions to use AEmotionStudio/magenta-realtime-2-mirror with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AEmotionStudio/magenta-realtime-2-mirror with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-audio", model="AEmotionStudio/magenta-realtime-2-mirror")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AEmotionStudio/magenta-realtime-2-mirror", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Magenta RealTime 2 — mirror (weights only)
Offline mirror of the pure-PyTorch magenta-community port of Google DeepMind's Magenta RealTime 2, bundled for the MAESTRO app. Weights only — the model code is vendored in-app (Apache-2.0).
Layout: base/, small/ (each config.json + codec_shapes.json +
model.safetensors) and a shared musiccoca/ style processor.
License: model weights CC-BY-4.0 © Google LLC — commercial use permitted,
attribution required, outputs are the user's. Upstream port © multimodalart /
magenta-community (Apache-2.0 code). Original: google/magenta-realtime-2.
Model tree for AEmotionStudio/magenta-realtime-2-mirror
Base model
google/magenta-realtime-2 Finetuned
magenta-community/magenta-realtime-2