howl
howl copied to clipboard
Wake word detection modeling toolkit for Firefox Voice, supporting open datasets like Speech Commands and Common Voice.
Hello, I am trying to prepare dataset as in instruction, but having problems with pydantic. Some of them I solved by using `from pydantic_settings import BaseSettings` instead of `from pydantic...
On more recent versions of pydantic, `pydantic.BaseSettings` got changed to `pydantic_settings.BaseSettings`
while runing training.run.stitch_vocab_samples gave a error after few 1-10 training.wav files generation File "/home/nsl/asr/howl/howl/utils/sphinx_keyword_detector.py", line 25, in detect result = phrase.segments(detailed=True) File "/home/nsl/.local/lib/python3.8/site-packages/pocketsphinx/__init__.py", line 134, in segments return [ TypeError:...
Dear Author, Thanks for sharing your package. However, I cannot stitch vocab samples. When I use `python -m training.run.stitch_vocab_samples --dataset-path "datasets/fire/positive"` ,I get this error:`RuntimeError: Error opening '/tmp/temp.wav': System error.`...
Trainer class that manages model training logics given a config
Hi, I'm trying to generate a custom dataset as instructed. After the filtering process, I get this error: Traceback (most recent call last): File "/home/boaz/miniconda3/envs/howlenv/lib/python3.9/runpy.py", line 197, in _run_module_as_main return...
Dear Author, Thanks for sharing your package. In your example to generate the data set, "fire" has 2 parts of data, positive and negative. What is the positive data? Was...
Hi, I've trained a model and now would like to convert it into weights for browser deployment like `hey_firefox.js`. Could you help me understand how that file was generated?
Bumps [torch](https://github.com/pytorch/pytorch) from 1.10.1 to 2.2.0. Release notes Sourced from torch's releases. PyTorch 2.2: FlashAttention-v2, AOTInductor PyTorch 2.2 Release Notes Highlights Backwards Incompatible Changes Deprecations New Features Improvements Bug fixes...