78 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
dscripka	4e77db111c	Bug fix CI/CD	2023-01-28 10:59:40 -05:00
dscripka	23acb7e2c6	Updated CI/CD configuration for linux	2023-01-28 10:49:19 -05:00
dscripka	2361045930	Removed extra optional dependencies to fix package build and upload	2023-01-28 10:45:36 -05:00
dscripka	e0d942d602	Changed how dependencies are handled	2023-01-28 10:23:00 -05:00
dscripka	01265f3f5d	Updated tag for next release	2023-01-28 10:00:17 -05:00
dscripka	0d90360172	Added docs for 'hey jarvis' model	2023-01-28 09:45:39 -05:00
dscripka	50c1d08ce4	Adjusted docs, added first release of new 'hey jarvis' model, small bug fixes in preparation for next release	2023-01-28 09:40:28 -05:00
dscripka	911f675e06	Merge pull request #10 from dscripka/false_positive_mining ... False positive mining	2023-01-24 20:03:58 -05:00
dscripka	35e1988b5c	Unit tests passsing	2023-01-24 09:57:02 -05:00
dscripka	3503c927bf	Merge branch 'main' into false_positive_mining ... Merge main into working branch to fix some broken commits	2023-01-24 08:31:08 -05:00
dscripka	772b61a04e	Merge pull request #9 from dscripka/voice_activity_detection ... Voice activity detection	2023-01-22 21:07:23 -05:00
dscripka	db351b0eec	Set explicit dtype	2023-01-22 21:04:17 -05:00
dscripka	e9ceed3de6	Adjusted tests to pass on Windows	2023-01-22 20:56:24 -05:00
dscripka	49a1e93043	Added missing .onnx model	2023-01-22 20:46:41 -05:00
dscripka	58dcab4c45	added missing file	2023-01-22 20:09:30 -05:00
dscripka	2edd4f547e	Added some scripts and functions to assist with collection false activations	2023-01-22 20:06:37 -05:00
dscripka	ac35230906	Integrated silero VAD	2023-01-18 08:01:25 -05:00
dscripka	b261abd9be	Merge pull request #8 from dscripka/speex_noise_suppression_v2 ... Integrate basic noise suppression with Speex.	2023-01-17 20:05:34 -05:00
dscripka	cef0b7a758	Update README.md to discuss noise suppression	2023-01-17 19:20:23 -05:00
dscripka	1841e7e8d9	adjusted setup.py, last fix	2023-01-17 18:51:10 -05:00
dscripka	173f5aa11a	adjusted setup.py, debugging	2023-01-17 18:41:45 -05:00
dscripka	e01288b2fd	adjusted setup.py, bugfix	2023-01-17 18:35:48 -05:00
dscripka	654a11968d	adjusted setup.py	2023-01-17 18:31:45 -05:00
dscripka	981c68fe76	typo	2023-01-16 17:36:58 -05:00
dscripka	dbb95944cf	Adjusted CI config and tests for windows	2023-01-16 17:35:04 -05:00
dscripka	792f148bde	Dynamic generation of install_requires to enable speexdsp_ns package installation	2023-01-16 16:53:57 -05:00
dscripka	d663ea70ba	Added first version of 'hey jarvis' model	2023-01-16 15:56:15 -05:00
dscripka	d060f75f97	Removed premature import	2023-01-16 15:51:04 -05:00
dscripka	06dac3355c	fix newline characters	2023-01-16 15:46:25 -05:00
dscripka	054f7578d0	Test coverage for model.py back to 100%	2023-01-16 15:37:06 -05:00
dscripka	4d9d549930	Tests passing	2023-01-15 21:44:35 -05:00
David Scripka	f870d4340d	Integrated Speex noise suppression as an optional pre-processing step, adjust data augmentation method.	2023-01-14 19:48:09 -05:00
dscripka	81badf5728	Added example script to save audio around an activation to aid in debugging and performance testing, added new data augmentation method using random noise	2023-01-08 13:41:07 -05:00
dscripka	216f0329f4	Added example script for saving model activations to WAV files	2023-01-07 12:28:35 -05:00
dscripka	c32de6573d	Added new functionality to mix_clips_batch function, fixed bug in model in model.py	2023-01-04 07:28:22 -05:00
dscripka	5d8816e1b3	Added trained model for wake word	2023-01-01 22:48:11 -05:00
dscripka	737e8891b9	flake8 fixes	2023-01-01 19:15:57 -05:00
dscripka	69afd1e3e3	Reworked method in metrics.py for false positive calculations, added example notebooks for performance calculations	2023-01-01 19:08:37 -05:00
dscripka	f3a8692bda	method now supporst numpy array inputs, added more data agmentation options for function	2022-12-31 22:19:59 -05:00
dscripka	a3b6985cba	Added link to HuggingFace spaces demo, renamed some test files	2022-12-30 13:10:37 -05:00
dscripka	af628f5a2c	version bump fixing broken release	2022-12-30 12:06:37 -05:00
dscripka	bc764acca7	Fixed bug in package build	2022-12-30 12:01:17 -05:00
David Scripka	144a19ccb6	Fixed license classifier	2022-12-29 17:12:15 -05:00
David Scripka	525d71dbe7	Few edits & typos	2022-12-29 13:43:20 -05:00
David Scripka	5e5f5168af	Setup CI/CD to build and publish package to pypi	2022-12-29 12:58:23 -05:00
dscripka	43a4616696	Fixes for later version of mypy	2022-12-23 20:40:15 -05:00
dscripka	f9125f9600	Adjusted model naming convention and added 2 new models	2022-12-23 17:02:56 -05:00
dscripka	8a96edb5e9	Tests passing	2022-12-23 16:39:26 -05:00
dscripka	bc3febc9c1	Final preparation for initial release	2022-12-23 14:51:49 -05:00
dscripka	7fd47574cc	More fluent speech commands experiments, updated data prep function	2022-12-21 22:03:19 -05:00