A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Improved CSKS with limited data using novel loss functions and transfer learning.
Synthetic speech data improves keyword spotting models with fewer real examples.
User-specific KWS system learns new keywords on-device.
A new multi-layer attention mechanism improves speech keyword recognition accuracy.
SpeechYOLO detects and locates speech objects in audio signals.
We consider multilingual bottleneck features (BNFs) for nearly zero-resource keyword spotting. This forms part of a United Nations effort using keyword spotting to support humanitarian relief programmes in parts of Africa where languages are severely under-resourced. We use 1920 isolated keywords (40 types, 34 minutes)…
Paper optimizes KWS models using NAS and quantization for limited resources.
Accurate on-device keyword spotting (KWS) with low false accept and false reject rate is crucial to customer experience for far-field voice control of conversational agents. It is particularly challenging to maintain low false reject rate in real world conditions where there is (a) ambient noise from external sources s…
Enhances KWS in vehicles with multi-source fusion.
We propose a practical approach based on federated learning to solve out-of-domain issues with continuously running embedded speech-based models such as wake word detectors. We conduct an extensive empirical study of the federated averaging algorithm for the "Hey Snips" wake word based on a crowdsourced dataset that mi…
The problem of keyword spotting i.e. identifying keywords in a real-time audio stream is mainly solved by applying a neural network over successive sliding windows. Due to the difficulty of the task, baseline models are usually large, resulting in a high computational cost and energy consumption level. We propose a new…
Channel pruning and weight binarization improve keyword spotting accuracy.
Paper introduces regularization for multi-head attention to spot keywords.
Large size models are implemented in recently ASR system to deal with complex speech recognition problems. The num- ber of parameters in these models makes them hard to deploy, especially on some resource-short devices such as car tablet. Besides this, at most of time, ASR system is used to deal with real-time problem …
Self-training with noisy student-teacher boosts keyword spotting accuracy.
Keyword spotting--or wakeword detection--is an essential feature for hands-free operation of modern voice-controlled devices. With such devices becoming ubiquitous, users might want to choose a personalized custom wakeword. In this work, we present DONUT, a CTC-based algorithm for online query-by-example keyword spotti…
Neural model detects phoneme boundaries from speech, outperforming baselines.
We present our first efforts in building an automatic speech recognition system for Somali, an under-resourced language, using 1.57 hrs of annotated speech for acoustic model training. The system is part of an ongoing effort by the United Nations (UN) to implement keyword spotting systems supporting humanitarian relief…
The problem of identifying voice commands has always been a challenge due to the presence of noise and variability in speed, pitch, etc. We will compare the efficacies of several neural network architectures for the speech recognition problem. In particular, we will build a model to determine whether a one second audio…
New KWS neural networks improve accuracy and power efficiency.
We propose a max-pooling based loss function for training Long Short-Term Memory (LSTM) networks for small-footprint keyword spotting (KWS), with low CPU, memory, and latency requirements. The max-pooling loss training can be further guided by initializing with a cross-entropy loss trained network. A posterior smoothin…
Detects adversarial examples in deep speech recognition.
New eGRU unit improves keyword spotting on ultra-low-power devices.
We compare features for dynamic time warping (DTW) when used to bootstrap keyword spotting (KWS) in an almost zero-resource setting. Such quickly-deployable systems aim to support United Nations (UN) humanitarian relief efforts in parts of Africa with severely under-resourced languages. Our objective is to identify aco…
We explore the application of end-to-end stateless temporal modeling to small-footprint keyword spotting as opposed to recurrent networks that model long-term temporal dependencies using internal states. We propose a model inspired by the recent success of dilated convolutions in sequence modeling applications, allowin…
We consider the task of building compact deep learning pipelines suitable for deployment on storage and power constrained mobile devices. We propose a unified framework to learn a broad family of structured parameter matrices that are characterized by the notion of low displacement rank. Our structured transforms admit…
Mobile app improves speech recognition of names with user feedback.
Using Intel's Loihi neuromorphic research chip and ABR's Nengo Deep Learning toolkit, we analyze the inference speed, dynamic power consumption, and energy cost per inference of a two-layer neural network keyword spotter trained to recognize a single phrase. We perform comparative analyses of this keyword spotter runni…
Research forecasts electricity spot prices using stochastic volatility models.
Personal VAD detects target speaker voice activity efficiently.
Paper shows continuous speech recognition with EEG features, no speech input.
Continuous speech recognition from brain activity without vocalization.
Improved speech recognition using EEG and video.
AV-CPL uses continuous pseudo-labels for AVSR combining labeled and unlabeled data.
Paper explores EEG-based speech recognition using transformers, showing faster training and better performance for smaller vocabularies.
End-to-end speech recognition using EEG without speech input.
Paper improves EEG-based speech recognition using CTC and beam search.
Study Fourier estimator for spot volatility with unbounded coefficients and jumps.
Improved speech recognition for voice assistants by analyzing speech data.
The paper proposes using CWT and STFT for training neural speech models.
Derives pricing formulas for perpetual futures contracts.
Human infants can discover words directly from unsegmented speech signals without any explicitly labeled data. In this paper, we develop a novel machine learning method called nonparametric Bayesian double articulation analyzer (NPB-DAA) that can directly acquire language and acoustic models from observed continuous sp…
A dynamic keyword selection model for topic modeling of tweets.
This paper proposes a new loss using short-time Fourier transform (STFT) spectra for the aim of training a high-performance neural speech waveform model that predicts raw continuous speech waveform samples directly. Not only amplitude spectra but also phase spectra obtained from generated speech waveforms are used to c…
CL methods improve monolingual ASR models across new tasks without forgetting past data.
This paper examines the speaker identification potential of breath sounds in continuous speech. Speech is largely produced during exhalation. In order to replenish air in the lungs, speakers must periodically inhale. When inhalation occurs in the midst of continuous speech, it is generally through the mouth. Intra-spee…
Automatic continuous speech recognition (CSR) is sufficiently mature that a variety of real world applications are now possible including large vocabulary transcription and interactive spoken dialogues. This paper reviews the evolution of the statistical modelling techniques which underlie current-day systems, specific…
Paper defends LSTM-based text classification models from backdoor attacks.