SOTAVerified

Keyword Spotting

In speech processing, keyword spotting deals with the identification of keywords in utterances.

( Image credit: Simon Grest )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 251–300 of 407 papers

TitleStatusHype
Keyword Transformer: A Self-Attention Model for Keyword SpottingCode1
Auto-KWS 2021 Challenge: Task, Datasets, and BaselinesCode1
Prototype-based Personalized Pruning—0
SubSpectral Normalization for Neural Audio Data Processing—0
EdgeCRNN: an edgecomputing oriented model of acoustic feature enhancement for keyword spotting—0
An Ultra-low Power RNN Classifier for Always-On Voice Wake-Up Detection Robust to Real-World Scenarios—0
The NPU System for the 2020 Personalized Voice Trigger ChallengeCode0
Meta-Learning for improving rare word recognition in end-to-end ASR—0
Dynamic curriculum learning via data parameters for noise robust keyword spotting—0
Query-by-Example Keyword Spotting system using Multi-head Attention and Softtriple Loss—0
Modular approach to data preprocessing in ALOHA and application to a smart industry use case—0
Speech Enhancement for Wake-Up-Word detection in Voice Assistants—0
Learning Efficient Representations for Keyword Spotting with Triplet LossCode1
Neural Networks for Keyword Spotting on IoT Devices—0
A 510-nW Wake-Up Keyword-Spotting Chip Using Serial-FFT-Based MFCC and Binarized Depthwise Separable CNN in 28-nm CMOS—0
EfficientNet-Absolute Zero for Continuous Speech Keyword SpottingCode1
Resource-efficient DNNs for Keyword Spotting using Neural Architecture Search and QuantizationCode0
Training Wake Word Detection with Synthesized Speech Data on Confusion Words—0
Optimize what matters: Training DNN-HMM Keyword Spotting Model Using End Metric—0
Decentralizing Feature Extraction with Quantum Convolutional Neural Network for Automatic Speech RecognitionCode1
RNNAccel: A Fusion Recurrent Neural Network Accelerator for Edge Intelligence—0
Deep Convolutional Spiking Neural Networks for Keyword Spotting—0
MicroNets: Neural Network Architectures for Deploying TinyML Applications on Commodity MicrocontrollersCode1
Small-Footprint Keyword Spotting with Multi-Scale Temporal ConvolutionCode1
Low-Power Low-Latency Keyword Spotting and Adaptive Control with a SpiNNaker 2 Prototype and Comparison with Loihi—0
Hardware Aware Training for Efficient Keyword Spotting on General Purpose and Specialized Hardware—0
AutoKWS: Keyword Spotting with Differentiable Architecture Search—0
Seeing wake words: Audio-visual Keyword SpottingCode1
Neural Architecture Search For Keyword Spotting—0
Howl: A Deployed, Open-Source Wake Word Detection SystemCode1
Learning Graph Edit Distance by Graph Neural Networks—0
WSRNet: Joint Spotting and Recognition of Handwritten Words—0
Neural ODE with Temporal Convolution and Time Delay Neural Networks for Small-Footprint Keyword SpottingCode0
Few-Shot Keyword Spotting With Prototypical NetworksCode1
BSL-1K: Scaling up co-articulated sign language recognition using mouthing cuesCode1
TERA: Self-Supervised Learning of Transformer Encoder Representation for SpeechCode1
Always-On, Sub-300-nW, Event-Driven Spiking Neural Network based on Spike-Driven Clock-Generation and Clock- and Power-Gating for an Ultra-Low-Power Intelligent Device—0
Exploring Filterbank Learning for Keyword Spotting—0
Training Keyword Spotting Models on Non-IID Data with Federated Learning—0
Metric Learning for Keyword Spotting—0
Multi-Task Network for Noise-Robust Keyword Spotting and Speaker Verification using CTC-based Soft VAD and Global Query Attention—0
Domain Aware Training for Far-field Small-footprint Keyword Spotting—0
Reformulating Information Retrieval from Speech and Text as a Detection Problem—0
Small-Footprint Open-Vocabulary Keyword Spotting with Quantized LSTM Networks—0
Phoneme Boundary Detection using Learnable Segmental FeaturesCode1
Training Keyword Spotters with Limited and Synthesized Speech DataCode2
Learning To Detect Keyword Parts And Whole By Smoothed Max Pooling—0
Performance-Oriented Neural Architecture Search—0
A Multi-oriented Chinese Keyword Spotter Guided by Text Line Detection—0
Predicting detection filters for small footprint open-vocabulary keyword spotting—0
Show:102550
← PrevPage 6 of 9Next →
All datasetsQUESSTGoogle Speech Commandshey SiriFKDGoogle Speech Commands V2 35TensorFlowVoxForgeGoogle Speech Commands (v2)Google Speech Commands V2 12

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1NNI non-filtered(for the development set)Cnxe6.09—Unverified
2NNI Choi(for the development set)Cnxe5.89—Unverified
3NTU rnn (eval)Cnxe2.01—Unverified
4NTU dtw (eval)Cnxe2.01—Unverified
5NTU dtw (dev)Cnxe2.01—Unverified
6NTU rnn (dev)Cnxe2.01—Unverified
7ELiRF SDTW (eval)Cnxe1.19—Unverified
8ELiRF SDTW-avg (eval)Cnxe1.07—Unverified
9ELiRF SDTW (dev)Cnxe1.07—Unverified
10CUNY [Subseq+MFCC] (eval)Cnxe1.07—Unverified
#ModelMetricClaimedVerifiedStatus
1WaveFormerGoogle Speech Commands V2 1298.8—Unverified
2QNNGoogle Speech Commands V2 3598.6—Unverified
3TripletLoss-res15Google Speech Commands V1 1298.56—Unverified
4M2DGoogle Speech Commands V2 3598.5—Unverified
5EAT-SGoogle Speech Commands V2 3598.15—Unverified
6Audio Spectrogram TransformerGoogle Speech Commands V2 3598.11—Unverified
7EdgeCRNN 2.0×Google Speech Commands V2 1298.05—Unverified
8BC-ResNet-8Google Speech Commands V1 1298—Unverified
9HTS-ATGoogle Speech Commands V2 3598—Unverified
10Wav2KWSGoogle Speech Commands V1 1297.9—Unverified
#ModelMetricClaimedVerifiedStatus
1Stacked 1D CNNError Rate1.99—Unverified
2End-to-end DNN-HMMError Rate1.7—Unverified
3HEiMDaLError Rate0.45—Unverified
#ModelMetricClaimedVerifiedStatus
1Res26Accuracy95.88—Unverified
2EfficientNet-A0 + SA + TLAccuracy95.83—Unverified
#ModelMetricClaimedVerifiedStatus
1QuaternionNeuralNetworkAccuracy (10-fold)98.53—Unverified
2SSAMBAAccuracy (10-fold)97.4—Unverified
#ModelMetricClaimedVerifiedStatus
1TensorFlow's model version 2TFMA89.7—Unverified
2TensorFlow's model version 1TFMA85.4—Unverified
#ModelMetricClaimedVerifiedStatus
12D-ConvNetAccuracy (%)95.4—Unverified
21D-ConvNetAccuracy (%)93.7—Unverified
#ModelMetricClaimedVerifiedStatus
1Quaternion Neural NetworksAccuracy(10-fold)98.53—Unverified
#ModelMetricClaimedVerifiedStatus
1MicroNet-KWS-LAccuracy95.3—Unverified