Voice Trigger Detection from Lvcsr Hypothesis Lattices Using Bidirectional Lattice Recurrent Neural Networks

Jeon, Woojay; Liu, Chengxi; Mason, H. Benjamin

doi:10.1109/icassp.2019.8682617

Cited by 10 publications

(14 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lattice embeddings are obtained by treating the lattice as a graph and processing it using multiple hidden layers of multi-headed self-attention operation. These embeddings have been shown to be highly informative for FTM task [12,1], but they can be obtained only by running full-fledged ASR on the audio which is expensive to be run on-device and invades user privacy in case of a false trigger. Moreover, the LatticeGNN model needs to be retrained if the distribution of the input lattice features changes due to any changes in the acoustic model, language model or the ASR decoding parameters.…”

Section: Latticegnn Ftm and Lattice Embeddingsmentioning

confidence: 99%

Knowledge Transfer for Efficient on-Device False Trigger Mitigation

Dighe

Marchi

Vishnubhotla

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we address the task of determining whether a given utterance is directed towards a voice-enabled smart-assistant device or not. An undirected utterance is termed as a "false trigger" and false trigger mitigation (FTM) is essential for designing a privacycentric non-intrusive smart assistant. The directedness of an utterance can be identified by running automatic speech recognition (ASR) and determining the user intent by analyzing the ASR transcript. Yet, in case of a false trigger, transcribing the audio using ASR itself is strongly undesirable. To alleviate this issue, we propose an LSTM-based FTM architecture which determines the user intent from acoustic features directly without explicitly generating ASR transcripts from the audio. The proposed models are smallfootprint and can be run on-device with limited computational resources. During training, the model parameters are optimized using a knowledge transfer approach where a more accurate self-attention graph neural network model [1] serves as the teacher. Given the whole audio snippets, our approach mitigates 87% of false triggers at 99% true positive rate (TPR), and in a streaming audio scenario, the system listens to only 1.69s of the false trigger audio before rejecting it while achieving the same TPR.

show abstract

Section: Latticegnn Ftm and Lattice Embeddingsmentioning

confidence: 99%

Knowledge Transfer for Efficient on-Device False Trigger Mitigation

Dighe

Marchi

Vishnubhotla

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Although there has been much work on sequential classification [22] and trigger word spotting [23], [24], [25] in the speech recognition literature, there have been few studies of sequential classification of continuous RF data time streams [26], [27]. Moreover, these studies focus exclusively on gross body motion classification.…”

Section: Sequential Classificationmentioning

confidence: 99%

Word-Level ASL Recognition and Trigger Sign Detection with RF Sensors

Rahman

Kurtoğlu

Mdrafi

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Current research in the recognition of American Sign Language (ASL) has focused on perception using video or wearable gloves. However, deaf ASL users have expressed concern about the invasion of privacy with video, as well as the interference with daily activity and restrictions on movement presented by wearable gloves. In contrast, RF sensors can mitigate these issues as it is a non-contact ambient sensor that is effective in the dark and can penetrate clothes, while only recording speed and distance. Thus, this paper investigates RF sensing as an alternative sensing modality for ASL recognition to facilitate interactive devices and smart environments for the deaf and hardof-hearing. In particular, the recognition of up to 20 ASL signs, sequential classification of signing mixed with daily activity, and detection of a trigger sign to initiate human-computer interaction (HCI) via RF sensors is presented. Results yield %91.3 ASL word-level classification accuracy, %92.3 sequential recognition accuracy, 0.93 trigger recognition rate.

show abstract

“…The voice trigger is similar to keyword spotting that detects one or more predefined keywords from a sequence of speech signals [31,32]. It is widely used to wake up personal assistant devices.…”

Section: Conventional Voice Triggermentioning

confidence: 99%

Front-End of Vehicle-Embedded Speech Recognition for Voice-Driven Multi-UAVs Control

Park

2020

Applied Sciences

View full text Add to dashboard Cite

For reliable speech recognition, it is necessary to handle the usage environments. In this study, we target voice-driven multi-unmanned aerial vehicles (UAVs) control. Although many studies have introduced several systems for voice-driven UAV control, most have focused on a general speech recognition architecture to control a single UAV. However, for stable voice-controlled driving, it is essential to handle the environmental conditions of UAVs carefully, including environmental noise that deteriorates recognition accuracy, and the operating scheme, e.g., how to direct a target vehicle among multiple UAVs and switch targets using speech commands. To handle these issues, we propose an efficient, vehicle-embedded speech recognition front-end for multi-UAV control via voice. First, we propose a noise reduction approach that considers non-stationary noise in outdoor environments. The proposed method improves the conventional minimum mean squared error (MMSE) approach to handle non-stationary noises, e.g., babble and vehicle noises. In addition, we propose a multi-channel voice trigger method that can control multiple UAVs while efficiently directing and switching the target vehicle via speech commands. We evaluated the proposed methods on speech corpora, and the experimental results demonstrate that the proposed methods outperform the conventional approaches. In trigger word detection experiments, our approach yielded approximately 7%, 12%, and 3% relative improvements over spectral subtraction, adaptive comb filtering, and the conventional MMSE, respectively. In addition, the proposed multi-channel voice trigger approach achieved approximately 51% relative improvement over the conventional approach based on a single trigger word.

show abstract

Voice Trigger Detection from Lvcsr Hypothesis Lattices Using Bidirectional Lattice Recurrent Neural Networks

Cited by 10 publications

References 8 publications

Knowledge Transfer for Efficient on-Device False Trigger Mitigation

Knowledge Transfer for Efficient on-Device False Trigger Mitigation

Word-Level ASL Recognition and Trigger Sign Detection with RF Sensors

Front-End of Vehicle-Embedded Speech Recognition for Voice-Driven Multi-UAVs Control

Contact Info

Product

Resources

About