audio.cpp
breeze-tts-2-audiocpp-q8-dataset
BreezeTTS 2 synthetic speech dataset
Generated with the prebuilt 0xShug0/audio.cpp v0.8.1 CUDA 12.8 bundle.
Samples: 100
Audio: 0.11 hours
Format: 24 kHz mono WAV
Engine: 0xShug0/audio.cpp
Version: v0.8.1 prebuilt CUDA 12.8
Family: breeze_tts
Model: BreezeTTS 2
Precision: Q8_0 GGUF
Backend: CUDA
Model package
audio-cpp/audio.cpp-gguf
Breeze-TTS-2-GGUF/breeze-tts-2-q8_0.gguf
Metadata
metadata.jsonl includes the original manifest fields plus… See the full description on the dataset page: https://huggingface.co/datasets/Shadow0482/breeze-tts-2-audiocpp-q8-dataset.audio-cpp-model-archives
audio.cpp model archives
Public archive cache for selected audio.cpp model folders.
Files are .tar.zst archives intended to be downloaded and extracted into /models on GPU instances.
mkdir -p /models
wget -O /workspace/OmniVoice.tar.zst https://huggingface.co/datasets/laixuanthoi/audio-cpp-model-archives/resolve/main/OmniVoice.tar.zst
tar --use-compress-program='zstd -d' -xf /workspace/OmniVoice.tar.zst -C /models
See sha256sums-omnivoice.txt for integrity checks.