Generate Any 3D Scene in Seconds
Google AI Studio Starter Apps
Code for the paper Language Models are Unsupervised Multitask Learners
Chinese text-to-speech engine
MANTI - Mastering Advanced N-Termini Interpretation
Learning to Act by Watching Unlabeled Online Videos
WaveRNN Vocoder + TTS
muda.pl - MQ unified data assembler
Optional scripts for hubot, opt in via hubot-scripts.json
Portuguese ASR model fine-tuned on XLSR-53 for 16kHz audio input