OpenMOSS/MOSS-TTS-Nano
Realtime TTS without GPU
View on index · View in 3D Map
Related repos
- Models for speech and sound generation
- self-hosted lightweight text-to-speech server with web UI and GPU support
- Handheld vector network analyzer with LCD display
- GPU kernels written from scratch
- TTS, STT, and STS on Apple Silicon
- Harness for training LLMs on a single GPU node
- Hardware-accelerated GPU terminal emulator
- Linux operating system for embedded devices
- Trains text-to-speech models with 2000x realtime generation on GPU
- competes to train GPT-2 models super fast on 8xH100 GPUs using speedrun benchmarks
