We present a novel deep Recurrent Neural Network (RNN) model for acoustic modelling in Automatic Speech Recognition (ASR). We term our contribution as a TC-DNN-BLSTM-DNN model, the model combines a Deep Neural Network (DNN) with Time Convolution (TC), followed by a Bidirectional Long Short-Term Memory (BLSTM), and a fi…
On-device research index
arXiv research
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.
169,181 papers · 148 categories
Trend · papers per month
48 results for “DNN acoustic models”
We study large-scale kernel methods for acoustic modeling and compare to DNNs on performance metrics related to both acoustic modeling and recognition. Measuring perplexity and frame-level classification accuracy, kernel-based acoustic models are as effective as their DNN counterparts. However, on token-error-rates DNN…
A new batch optimisation framework using Natural Gradient improves DNN acoustic models.
problem Optimizing DNN acoustic models for better word error rate approximation.
method Proposes a Natural Gradient (NG) approach to sequence training, correcting the gradient based on local curvature of KL-divergence.
result The NG method converges more quickly and can be applied to any sequence discriminative training criterion.
Deep neural networks (DNNs) are now a central component of nearly all state-of-the-art speech recognition systems. Building neural network acoustic models requires several design decisions including network architecture, size, and training loss function. This paper offers an empirical investigation on which aspects of …
Speaker-Independent Speech-Driven Visual Speech Synthesis using Domain-Adapted Acoustic Modelseess.AS
Improved visual speech synthesis using adapted ASR acoustic models.
problem Lack of synchronized audio, video, and depth data for speaker-independent speech-driven visual speech synthesis.
method Adapted an ASR acoustic model trained on audio-only data to the visual speech synthesis domain.
result Viewers significantly prefer animations generated from the adapted ASR acoustic model.
Improved acoustic modeling with attentive adversarial learning.
problem Domain variability in acoustic modeling.
method Proposes an attentive ADIT method with an attention mechanism to improve domain-invariance of deep features.
result Improves deep feature domain-invariance and senone-discriminativity over ADIT.
Exploiting Low-dimensional Structures to Enhance DNN Based Acoustic Modeling in Speech Recognitioncs.CL
We propose to model the acoustic space of deep neural network (DNN) class-conditional posterior probabilities as a union of low-dimensional subspaces. To that end, the training posteriors are used for dictionary learning and sparse coding. Sparse representation of the test posteriors using this dictionary enables proje…
Deep neural networks (DNNs) have been successfully applied to a wide variety of acoustic modeling tasks in recent years. These include the applications of DNNs either in a discriminative feature extraction or in a hybrid acoustic modeling scenario. Despite the rapid progress in this area, a number of challenges remain …
We have recently shown that deep Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) outperform feed forward deep neural networks (DNNs) as acoustic models for speech recognition. More recently, we have shown that the performance of sequence trained context dependent (CD) hidden Markov model (HMM) acoustic m…
Deep Neural Network (DNN) acoustic models have yielded many state-of-the-art results in Automatic Speech Recognition (ASR) tasks. More recently, Recurrent Neural Network (RNN) models have been shown to outperform DNNs counterparts. However, state-of-the-art DNN and RNN models tend to be impractical to deploy on embedde…
We study large-scale kernel methods for acoustic modeling in speech recognition and compare their performance to deep neural networks (DNNs). We perform experiments on four speech recognition datasets, including the TIMIT and Broadcast News benchmark tasks, and compare these two types of models on frame-level performan…
Spatial Diffuseness Features for DNN-Based Speech Recognition in Noisy and Reverberant Environmentscs.CL
We propose a spatial diffuseness feature for deep neural network (DNN)-based automatic speech recognition to improve recognition accuracy in reverberant and noisy environments. The feature is computed in real-time from multiple microphone signals without requiring knowledge or estimation of the direction of arrival, an…
NLE embeds labels for domain adaptation with neural networks.
problem Adapting deep neural networks with unpaired source and target domain data.
method Distill source-domain knowledge into l-vectors, soft targets for adaptation.
result 14.1% relative word error rate reduction over direct re-training.
Two methods use DNN-HMM for global SNR estimation of speech signals.
problem Estimating global SNR of speech signals in various noise conditions.
method Dropout approximation for uncertainty estimation and noise-specific regressors.
result Improved SNR estimation accuracy compared to existing methods.
Proposes COALA method for learning audio representations aligned with tags.
problem Lack of annotated data for high-performance audio representation learning.
method Aligns latent representations of audio and tags using a contrastive loss.
result Audio embedding model captures both acoustic and semantic characteristics.
In this paper we study the probabilistic properties of the posteriors in a speech recognition system that uses a deep neural network (DNN) for acoustic modeling. We do this by reducing Kaldi's DNN shared pdf-id posteriors to phone likelihoods, and using test set forced alignments to evaluate these using a calibration s…
In this paper we describe the recent advancements made in the IBM i-vector speaker recognition system for conversational speech. In particular, we identify key techniques that contribute to significant improvements in performance of our system, and quantify their contributions. The techniques include: 1) a nearest-neig…
ASA improves ASR by adapting SD models to SI model's deep feature distribution.
problem Improving ASR performance on new speakers with limited data.
method Adversarial learning to regularize SD model's deep features to match SI model's.
result ASA achieves significant word error rate improvements over SI models.
Long Short-Term Memory (LSTM) is a recurrent neural network (RNN) architecture that has been designed to address the vanishing and exploding gradient problems of conventional RNNs. Unlike feedforward neural networks, RNNs have cyclic connections making them powerful for modeling sequences. They have been successfully u…
This paper presents sampling-based speech parameter generation using moment-matching networks for Deep Neural Network (DNN)-based speech synthesis. Although people never produce exactly the same speech even if we try to express the same linguistic and para-linguistic information, typical statistical speech synthesis pr…
Paper introduces a new optimisation method combining NG and Hessian Free for sequence training.
problem Overfitting and inefficiency in training DNNs with sequence criteria.
method Combines Natural Gradient and Hessian Free methods for better path traversal on parameter manifold.
result Achieves larger reductions in WER and lower WER compared to NG and HF methods.
New task AQA tackles acoustic reasoning from sound scenes.
problem Promote research in acoustic reasoning.
method Generate acoustic scenes from elementary sounds and formulate questions.
result Preliminary results with models FiLM and MAC show promise.
Bayesian SHMM discovers acoustic units from unlabeled speech.
problem Discovering language-specific acoustic units from unlabeled speech.
method Bayesian Subspace Hidden Markov Model (SHMM) trained on labeled data to find new acoustic units on target language.
result Significantly outperforms previous HMM-based systems and compares favorably with Variational Auto Encoder-HMM.
Paper aims to find joint representation between vocal tract geometry and speech sound acoustics.
problem Finding a joint latent representation between articulatory and acoustic domains for vowel sounds.
method Invertible neural network models, convolutional autoencoder, normalizing flows, semi-supervised learning.
result Satisfactory performance in articulatory-to-acoustic and acoustic-to-articulatory mapping.
This paper compares new speech synthesis methods and finds Wavenet vocoders and AR models perform best.
problem Improving speech synthesis quality using advanced machine learning techniques.
method Large-scale crowdsourced evaluation of vocoding and acoustic modeling techniques.
result Wavenet vocoders and AR models outperform conventional methods in speech synthesis quality.
Paper predicts EEG features from acoustic features using RNN and GAN.
problem Predicting EEG features from acoustic features.
method Recurrent Neural Network (RNN) and Generative Adversarial Network (GAN).
result Lower RMSE and normalized RMSE values compared to generating acoustic features from EEG features.
We present a supervised neural network model for polyphonic piano music transcription. The architecture of the proposed model is analogous to speech recognition systems and comprises an acoustic model and a music language model. The acoustic model is a neural network used for estimating the probabilities of pitches in …
Convolutional Neural Networks (CNNs) have proven very effective in image classification and show promise for audio. We use various CNN architectures to classify the soundtracks of a dataset of 70M training videos (5.24 million hours) with 30,871 video-level labels. We examine fully connected Deep Neural Networks (DNNs)…
The Acoustic Camouflage Phenomenon: Re-evaluating Speech Features for Financial Risk Predictioncs.SD
Study shows integrating acoustic features in financial forecasting models can degrade performance.
problem Predicting stock market volatility from corporate earnings calls using speech features.
method Empirical investigation of acoustic feature extraction in teleconference environments using a two-stream late-fusion architecture.
result Integrating acoustic features via late fusion significantly degraded performance, reducing recall to 47.08%.
Enhances KWS in vehicles with multi-source fusion.
problem Improving precision and recall rates in vehicle keyword spotting.
method Integrates vehicle information into a DNN for speech classification and selects optimal sensitivity parameters.
result Significantly improved performance metrics (precision, recall, MSE) compared to baseline.
Improved multi-speaker TTS using GANs and waveform loss.
problem Training acoustic models for neural vocoders in multi-speaker TTS systems.
method Proposed frameworks incorporating Wasserstein GAN with gradient penalty (WGAN-GP) and discretized mixture logistic loss (DML) into acoustic models trained with WaveNet.
result Acoustic models trained with WGAN-GP and DML loss achieve highest subjective evaluation scores in multi-speaker TTS.
This paper introduces a model of environmental acoustic scenes which adopts a morphological approach by ab-stracting temporal structures of acoustic scenes. To demonstrate its potential, this model is employed to evaluate the performance of a large set of acoustic events detection systems. This model allows us to expli…
Improved hybrid acoustic model using interleaved self-attention and convolution.
problem Limited application of transformer in hybrid acoustic models.
method Proposed a model structure with interleaved self-attention and 1D convolution.
result Competitive recognition results on Librispeech dataset.
Three LF training criteria improve neural network acoustic models without cross-entropy pre-training.
problem Improving purely sequence-trained neural network acoustic models.
method Comparison of three lattice-free discriminative training criteria (MMI, bMMI, sMBR) on LVCSR tasks.
result LF-bMMI models outperform plain LF-MMI models by 5% WER on Switchboard datasets.
This work learns shared word embeddings for acoustic and phonetic sequences.
problem Mapping variable-length acoustic and phonetic sequences to fixed-dimensional vectors.
method Weak supervision and binary classification task to predict word similarity.
result Best model achieves an F1 score of 0.95 for binary classification.
This paper improves speech recognition by using raw waveform signals in multi-span CNN acoustic models.
problem Improving speech recognition accuracy using raw waveform signals.
method Proposes a novel multi-span structure for acoustic modelling based on raw waveform signals with multiple CNN input layers.
result Multi-span acoustic models yield a lower word error rate (WER) than traditional FBANK feature-based models.
Growing interest in automatic speaker verification (ASV)systems has lead to significant quality improvement of spoofing attackson them. Many research works confirm that despite the low equal er-ror rate (EER) ASV systems are still vulnerable to spoofing attacks. Inthis work we overview different acoustic feature spaces…
Geometry of Music Perceptionmath.DG
Geometric model explains music perception combining neuroscience and acoustics.
problem Rationalize and predict psycho-acoustic phenomena in music perception.
method Combining neuroscientific theories with acoustic observations, a geometric model of the space of all chords is created.
result The geometric model allows for rigorous studies of psychoacoustic quantities like roughness and harmonicity.
CLEAR dataset for acoustic reasoning tasks.
problem Acoustic reasoning and question answering.
method Data generation from elementary sounds, functional programs for question composition.
result Validation of current state-of-the-art visual Q&A models on AQA task.
U-Net trained to recover acoustic interference striations from distorted data.
problem Recovering acoustic interference striations from distorted signals.
method Training a U-Net using a random mode-coupling matrix model to generate training data.
result U-Net successfully recovers AISs under various conditions.
An improved algorithm for acoustic model parameter estimation.
problem Acoustic physical model parameter estimation for sound design.
method Multi-stage algorithm with deep learning, heuristics, and stochastic optimization.
result Optimization method refines deep learning estimates and improves objective metrics.
Improved ASR for English-isiZulu code-switched speech with semi-supervised training.
problem Improving ASR for code-switched speech between English and isiZulu.
method Semi-supervised training using automatic transcription of multilingual speech data.
result Semi-supervised training achieved significant WER reduction in ASR performance.
Acoustic Neighbor Embeddingseess.AS
Acoustic Neighbor Embeddings map speech and text to fixed dimensions for phonetic confusability.
problem Mapping speech and text to fixed dimensions for phonetic confusability.
method Adapting SNE to sequential inputs, training two encoder neural networks.
result More accurate results with low-dimensional embeddings in word recognition tasks.
Paper proposes cost-sensitive detection for environmental acoustic sensing.
problem Infeasibility of manual analysis for large-scale acoustic data.
method Cost-sensitive classification with variational autoencoders in Neyman-Pearson framework.
result Improved control over false positive and false negative rates.
Unified framework for speaker-adaptive models using scaling and bias codes.
problem Improving speaker-adaptive neural-network based speech synthesis systems.
method Unified framework representation and generalized scaling and bias codes.
result Improves performance of speaker adaptation compared to conventional methods.
Paper proposes a voting method to improve acoustic scene classification.
problem Improving acoustic scene classification accuracy.
method Punishment voting algorithm based on super categories construction.
result Punishment voting significantly improves classification performance.
Enhances sound texture in CNN for better acoustic scene classification.
problem Limited understanding of how CNNs perceive audio scenes.
method Used Class Activation Mapping (CAM) to analyze log-Mel features and proposed edge enhancement using DoG and Sobel operators.
result Edge-enhanced log-Mel features improve CNN performance in acoustic scene classification.
ConvNet classifies whale vocalizations and ambient noise in acoustic recordings.
problem Automated detection and classification of marine mammal vocalizations in acoustic recordings.
method Convolutional Neural Network with a novel acoustic representation.
result Classifier accurately detects and classifies whale vocalizations and ambient noise.