Automatic Speech Recognition
Transformers
qwen3-asr
latent-reasoning
test-time-compute
parameter-efficient
Instructions to use voidful/latentASR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use voidful/latentASR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="voidful/latentASR")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("voidful/latentASR", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| [project] | |
| name = "latentASR" | |
| version = "0.1.0" | |
| description = "Continuous latent test-time scaling for frozen Qwen3-ASR backbones." | |
| readme = "README.md" | |
| requires-python = ">=3.10" | |
| license = { text = "Apache-2.0" } | |
| authors = [ | |
| { name = "LatentASR contributors" } | |
| ] | |
| dependencies = [ | |
| "torch>=2.6", | |
| "torchaudio>=2.6", | |
| "transformers>=4.57", | |
| "datasets>=3.6", | |
| "huggingface-hub>=0.34", | |
| "qwen-asr>=0.0.6", | |
| "qwen-omni-utils>=0.0.8", | |
| "jiwer>=4.0", | |
| "peft>=0.18", | |
| "numpy>=1.24", | |
| "tqdm>=4.66" | |
| ] | |
| [tool.setuptools] | |
| py-modules = [ | |
| "config", | |
| "data", | |
| "eval", | |
| "losses", | |
| "model", | |
| "peft_utils", | |
| "train", | |
| "utils" | |
| ] | |