#
kws
Here are 27 public repositories matching this topic...
A collection of resources to make a smart speaker
-
Updated
Dec 20, 2019
wake word engine benchmark framework
benchmark
benchmark-framework
voice-control
hotword-detection
keyword-spotting
voice-assistant
wake-word-detection
kws
wakeword
-
Updated
Nov 25, 2021 - Python
Few-shot Keyword Spotting in Any Language and Multilingual Spoken Word Corpus
speech-recognition
keyword-spotting
wake-word-detection
query-by-example
kws
keyword-search
few-shot-learning
-
Updated
Apr 6, 2022 - Jupyter Notebook
Cortex M KWS example with Tengine Lite.
-
Updated
Jul 19, 2020 - C
Wav2Keyword is keyword spotting(KWS) based on Wav2Vec 2.0. This model shows state-of-the-art in Speech commands dataset V1 and V2.
-
Updated
May 24, 2022 - Python
In this repository, we explore using a hybrid system consisting of a Convolutional Neural Network and a Support Vector Machine for Keyword Spotting task.
-
Updated
May 26, 2022 - Python
machine-learning
chatbot
nala
voice
voice-commands
voice-recognition
voice-chat
voicetext
pocketsphinx
voice-control
snowboy
voice-assistant
voice-activity-detection
voice-synthesis
kws
voice-computing
multiple-intents
voice-dataset
voice-datasets
picovoice
-
Updated
Aug 23, 2020 - Python
Keyword Search Recipe for Subword ASR
-
Updated
Jul 12, 2019 - Shell
Speech command recognition with capsule network & various NNs / KWS on Google Speech Command Dataset.
speech-recognition
keyword-spotting
capsule-networks
kws
speech-command-recognition
google-speech-command-dataset
-
Updated
Jan 28, 2019 - Python
Tensorflow implementation of "Small-Footprint Keyword Spotting with Multi-Scale Temporal Convolution"(INTERSPEECH 2020)
-
Updated
Nov 11, 2020 - Python
Keyword spotting for audio with attention (KWS model for audio)
-
Updated
Jul 15, 2021 - Jupyter Notebook
Models and training scripts for "LSTMs for Keyword Spotting with ReRAM-based Compute-In-Memory Architectures" (ISCAS 2021).
-
Updated
Mar 25, 2021 - Shell
Recognition of 38 speech commands in russian. Based on Yandex Cup 2021 ML Challenge: ASR
voice-commands
pytorch
hotword-detection
keyword-spotting
wake-word-detection
onnx
kws
russian-language
trigger-word-detection
speech-command-recognition
-
Updated
Dec 22, 2021 - Python
keyword-spotting
distillation
kws
teacher-student-learning
distillation-loss
soft-labels-distillation
-
Updated
Dec 19, 2021 - Python
TVM on Azure Sphere Platform
-
Updated
Jun 27, 2020 - C
Detect a set of predefined keywords in an audioclip.
-
Updated
Sep 23, 2020 - Jupyter Notebook
A collection of resources to make a smart speaker
-
Updated
Nov 14, 2018
Attention Implementation for KWS
-
Updated
Dec 17, 2021 - Jupyter Notebook
A KWS model trained on SpeechCommands dataset, written in PyTorch.
-
Updated
Nov 13, 2020 - Python
A work around for the Microsoft Azure Cognitive Services Keyword Spotting for UWP Apps
uwp
uwp-apps
xamarin-forms
keyword
workaround
microsoft-cognitive-services
azure-cognitive-services
cognitive-services
keyword-spotting
spotting
kws
xamarin-uwp
-
Updated
Dec 7, 2019 - C#
A node-red integration for https://github.com/mathquis/node-personal-wakeword
-
Updated
Oct 20, 2021 - JavaScript
Probabilistic relevance scores from PHOC embeddings
-
Updated
Feb 8, 2019 - Shell
Improve this page
Add a description, image, and links to the kws topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the kws topic, visit your repo's landing page and select "manage topics."
目前的多音字使用 pypinyin 或者 g2pM，精度有限，想做一个基于 BERT (或者 ERNIE) 多音字预测模型，简单来说就是假设某语言有 100 个多音字，每个多音字最多有 3 个发音，那么可以在 BERT 后面接 100 个 3 分类器（简单的 fc 层即可），在预测时，找到对应的分类器进行分类即可。
参考论文：
tencent_polyphone.pdf
数据可以用 https://github.com/kakaobrain/g2pM 提供的数据
进阶：多任务的 BERT
![image](https://user-images.githubusercontent.com/24568452