Text-to-Speech
ESPnet
PyTorch
English
audio
speech
instruction-following
singing-voice-synthesis
speech-language-model
Instructions to use espnet/bagpiper-tts-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ESPnet
How to use espnet/bagpiper-tts-sft with ESPnet:
from espnet2.bin.tts_inference import Text2Speech model = Text2Speech.from_pretrained("espnet/bagpiper-tts-sft") speech, *_ = model("text to generate speech from") - Notebooks
- Google Colab
- Kaggle
Download inference.yaml from espnet/bagpiper-tts-sft: direct link, hf CLI and curl.
- Browser
- Download file 220 Bytes
-
https://huggingface.co/espnet/bagpiper-tts-sft/resolve/main/inference.yaml
- Command line
-
hf download hf://espnet/bagpiper-tts-sft/inference.yaml
-
curl -L -o inference.yaml https://huggingface.co/espnet/bagpiper-tts-sft/resolve/main/inference.yaml
220 Bytes
| dtype: bfloat16 | |
| num_hypo: 1 | |
| enforce_modality: ["text", "audio"] | |
| audio: | |
| temperature: 0.8 | |
| topk: 20 | |
| cfg: 3 | |
| max_step: 2048 | |
| min_step: 50 | |
| text: | |
| temperature: 0.6 | |
| topk: 20 | |
| cfg: 1 | |
| max_step: 2048 | |
| min_step: 1 | |