# Piper TTS GGUF

Native audio.cpp GGUF conversion of the official Piper
[`en_US-lessac-medium`](https://huggingface.co/rhasspy/piper-voices/tree/main/en/en_US/lessac/medium)
voice. The package preserves the original FP32 tensors and embeds the
`piper_tts` model specification and voice configuration.

```bash
audiocpp_cli --task tts --family piper_tts \
  --model piper-en-us-lessac-medium-orig.gguf --backend cpu \
  --text "Hello from Piper." --out piper.wav
```

The archived Piper runtime source is MIT licensed; see `LICENSE`. The voice was
trained on the Lessac Blizzard 2013 dataset. Review the upstream model card and
dataset license before redistribution.