VoiceFilter-Lite separates speech from background in real-time for on-device speech recognition.
problem Separate speech from background in real-time for on-device speech recognition.
method Asymmetric loss, adaptive runtime suppression, quantization to 8-bit.
result VoiceFilter-Lite achieves real-time speech separation and maintains speech recognition performance.
Streaming ASR with transformer achieves low WER.
problem Real-time ASR with speech recognition.
method Time-restricted self-attention and triggered attention mechanisms.
result 2.8% and 7.2% WER for LibriSpeech test data.
Personal VAD detects target speaker voice activity efficiently.
problem Efficiently detect target speaker voice activity for reduced computational cost and battery usage.
method Trains a neural network conditioned on speaker embedding or verification score, outputs probabilities for three speech classes.
result Trained model with 130K parameters outperforms combined standard VAD and speaker recognition networks.
Streamable model improves speech recognition performance.
problem Improving speech recognition in long-form audio.
method Chunk-based attention model for streaming speech recognition.
result Streamable model maintains competitive performance and generalizes well to long-form speech.
CAT toolkit combines hybrid and E2E approaches for efficient speech recognition.
problem Improving speech recognition efficiency and latency.
method CTC-CRF based framework with contextualized soft forgetting.
result CAT achieves state-of-the-art results with simpler training and streaming ASR.
Improved multilingual speech recognition with low latency for nine Indic languages.
problem Imbalance in training data across languages and low latency in interactive applications.
method Conditioning on a language vector and training language-specific adapter layers.
result Lower word error rate than monolingual E2E models and conventional systems.
End-to-end speech recognition system trained on GPUs and CPUs.
problem Building state-of-the-art speech recognition systems.
method Utilizes CPUs and GPUs for training, data augmentation, and neural network updates. Uses vocal tract length perturbation and acoustic simulator for data augmentation. Employed Horovod allreduce for training.
result Achieved 7.92% WER on proprietary English Bixby open domain test set using a Bidirectional Full Attention (BFA) model.
We replace the Hidden Markov Model (HMM) which is traditionally used in in continuous speech recognition with a bi-directional recurrent neural network encoder coupled to a recurrent neural network decoder that directly emits a stream of phonemes. The alignment between the input and output sequences is established usin…
Improved speech recognition using EEG and video.
problem Enhancing continuous speech recognition systems.
method Implemented a CTC-based ASR model using EEG features.
result EEG features improve continuous visual speech recognition.
In this paper we demonstrate end-to-end continuous speech recognition (CSR) using electroencephalography (EEG) signals with no speech signal as input. An attention model based automatic speech recognition (ASR) and connectionist temporal classification (CTC) based ASR systems were implemented for performing recognition…
Continuous speech recognition from brain activity without vocalization.
problem Recognizing silent speech from EEG signals.
method Implemented a CTC ASR model using EEG signals.
result Demonstrated feasibility of EEG for continuous silent speech recognition.
The performance of automatic speech recognition systems(ASR) degrades in the presence of noisy speech. This paper demonstrates that using electroencephalography (EEG) can help automatic speech recognition systems overcome performance loss in the presence of noise. The paper also shows that distillation training of auto…
AV-CPL uses continuous pseudo-labels for AVSR combining labeled and unlabeled data.
problem Improving AVSR performance with labeled and unlabeled data.
method Semi-supervised method using continuous pseudo-labels generated by the same AVSR model.
result Significant improvements in VSR performance on LRS3 dataset.
Paper explores EEG-based speech recognition using transformers, showing faster training and better performance for smaller vocabularies.
problem Continuous speech recognition using EEG features.
method Transformer-based ASR model compared to RNN-based models.
result Transformer models perform better for smaller vocabularies but RNN models outperform them for larger vocabularies.
Paper uses surprisal to dynamically allocate computation between fast and slow models.
problem Dynamic allocation of computation in neural networks.
method Surprisal-based dynamic model selection.
result Model can match baseline performance with 15% fewer FLOPs.
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.
In this paper we demonstrate continuous noisy speech recognition using connectionist temporal classification (CTC) model on limited Chinese vocabulary using electroencephalography (EEG) features with no speech signal as input and we further demonstrate single CTC model based continuous noisy speech recognition on limit…
Improved speech emotion recognition using pre-trained language models.
problem Challenging task of speech emotion recognition for natural human-machine interaction.
method Fine-tuning pre-trained language models for text emotion recognition, combining with speech emotion recognition.
result 73.5% accuracy in speech emotion recognition on a subset of IEMOCAP dataset.
Paper improves EEG-based speech recognition using CTC and beam search.
problem Improving continuous speech recognition from EEG signals.
method Implemented CTC-based ASR system, initialized weights, used external language model, studied articulatory feature prediction.
result Enhanced performance of EEG-based speech recognition systems.
End-to-end model detects articulatory features from speech data.
problem Detecting articulatory features from speech data for various applications.
method Apply Listen, Attend and Spell (LAS) architecture and attention models.
result End-to-end training of manners and places of articulation detectors.
Large receptive field CNNs improve distant speech recognition.
problem Degrading performance of ASR systems in noisy environments.
method Investigated large receptive field CNN variants including recursive, dilated, and hourglass networks.
result Stacked hourglass networks show significant improvements in distant speech recognition.
Deep architecture learns transferable features for robust speech emotion recognition.
problem Robust and discriminative features for diverse speech emotion domains.
method Jointly uses CNN for domain-shared features and LSTM for domain-specific emotion classification.
result Transferable features provide gains up to 18.4% in speech emotion recognition.
New dataset for evaluating speech recognition fairness across demographics.
problem Lack of fairness metrics in speech recognition datasets.
method Developed Fair-Speech dataset with diverse demographic information.
result Helps evaluate ASR models for fairness across demographics.
A new deep neural network improves short-speech recognition.
problem Improving speaker recognition for short-duration speech segments.
method Proposes UtterIdNet, a novel deep neural network architecture.
result Consistent and stable performance for short segments, especially sub-second durations.
Survey on DNNs for speech processing, focusing on limited data challenges.
problem Challenges in training DNNs for speech tasks with limited data.
method Overview of techniques for few-shot speech processing.
result Promising few-shot techniques for speech processing.
Spoken language translation (SLT) has become very important in an increasingly globalized world. Machine translation (MT) for automatic speech recognition (ASR) systems is a major challenge of great interest. This research investigates that automatic sentence segmentation of speech that is important for enriching speec…
Study proposes a decision tree for more accurate depression recognition in speech.
problem Subjective bias in traditional depression diagnosis methods.
method A novel speech segment fusion method based on decision tree.
result The proposed decision tree model improves depression classification performance.
New robustness certificates for streaming models with a sliding window.
problem Applying robustness certificates to streaming data with correlated inputs.
method Deriving robustness certificates for models using a sliding window over a sequence of potentially correlated inputs.
result Guarantees hold for the average model performance across the entire stream, independent of stream size.
A new multi-layer attention mechanism improves speech keyword recognition accuracy.
problem Inaccurate attention weights in LSTM networks for speech keyword recognition.
method Introducing information from layers prior to feature extraction into attention weights calculations.
result The proposed multi-layer attention mechanism leads to more accurate attention weights and improved keyword spotting performance.
Mobile app improves speech recognition of names with user feedback.
problem Improving speech recognition of proper names on mobile devices.
method Data synthesis and user feedback for keyword-dependent metrics.
result User feedback can significantly improve speech recognition of names.
Quaternion neural networks improve distant speech recognition.
problem Challenges in distant speech recognition due to noise and reverberation.
method Quaternion neural networks process multi-channel audio signals as quaternion entities, capturing internal and external dependencies.
result QLSTM outperforms real-valued LSTM on multi-channel distant speech recognition tasks.
Paper proposes an online speech recognition model using Transformer.
problem Challenges in deploying Transformer-based E2E ASR for online speech recognition.
method Chunk self-attention encoder (chunk-SAE) and monotonic truncated attention (MTA) based self-attention decoder (SAD).
result Achieved 23.66% CER with 320 ms latency, significant improvement over offline models.
Speech recognition systems have achieved high recognition performance for several tasks. However, the performance of such systems is dependent on the tremendously costly development work of preparing vast amounts of task-matched transcribed speech data for supervised training. The key problem here is the cost of transc…
Recurrent neural network (RNN) language models (LMs) and Long Short Term Memory (LSTM) LMs, a variant of RNN LMs, have been shown to outperform traditional N-gram LMs on speech recognition tasks. However, these models are computationally more expensive than N-gram LMs for decoding, and thus, challenging to integrate in…
ADReSS Challenge at INTERSPEECH 2020 benchmarks speech recognition for Alzheimer's dementia.
problem Automated recognition of Alzheimer's dementia from spontaneous speech.
method Provides a benchmark speech dataset, defines two tasks (classification and regression), and presents baseline models.
result Demonstrates the feasibility of automated speech recognition for Alzheimer's dementia.
Attention-based encoder-decoder architectures such as Listen, Attend, and Spell (LAS), subsume the acoustic, pronunciation and language model components of a traditional automatic speech recognition (ASR) system into a single neural network. In previous work, we have shown that such architectures are comparable to stat…
In this work we explored building automatic speech recognition models for transcribing doctor patient conversation. We collected a large scale dataset of clinical conversations (14,000 hr), designed the task to represent the real word scenario, and explored several alignment approaches to iteratively improve data qua…
Novel bio-inspired masking for robust speech emotion recognition.
problem Noise degradation in speech emotion recognition.
method Cochlear cepstrogram-based contrastive learning with temporal and frequency masking.
result Improved speech emotion recognition performance on K-EmoCon benchmark.
Introspects convolutional speech recognition models using Gradient-adjusted Neuron Activation Profiles.
problem Lack of interpretability in deep learning ASR models.
method Gradient-adjusted Neuron Activation Profiles (GradNAPs) for feature and representation visualization.
result Gains insight into how data is processed in convolutional ASR models.
Mobile training improves speech recognition for users with unique speech characteristics.
problem Limited generalization of speaker-independent speech recognition models for users with very different speech characteristics.
method Securely training personalized end-to-end speech recognition models on mobile devices, splitting gradient computation to reduce memory usage.
result On-device personalization achieved 58.1% relative word error rate reduction compared to 63.7% in a server environment, with 18.7% performance degradation.
This paper proposes a Convolutional Neural Network (CNN) inspired by Multitask Learning (MTL) and based on speech features trained under the joint supervision of softmax loss and center loss, a powerful metric learning strategy, for the recognition of emotion in speech. Speech features such as Spectrograms and Mel-freq…
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 improves speech recognition by distilling knowledge from acoustic models.
problem Improving speech recognition accuracy using ensemble models.
method Proposes multi-teacher distillation strategies for joint CTC-attention end-to-end ASR systems, integrating error rate metric for optimization.
result Reports state-of-the-art error rates on various datasets and languages.
Although highly correlated, speech and speaker recognition have been regarded as two independent tasks and studied by two communities. This is certainly not the way that people behave: we decipher both speech content and speaker traits at the same time. This paper presents a unified model to perform speech and speaker …
Convolutional Neural Networks (CNNs) are effective models for reducing spectral variations and modeling spectral correlations in acoustic features for automatic speech recognition (ASR). Hybrid speech recognition systems incorporating CNNs with Hidden Markov Models/Gaussian Mixture Models (HMMs/GMMs) have achieved the …
In training speech recognition systems, labeling audio clips can be expensive, and not all data is equally valuable. Active learning aims to label only the most informative samples to reduce cost. For speech recognition, confidence scores and other likelihood-based active learning methods have been shown to be effectiv…
Enhances speech emotion recognition by adapting to varying time scales.
problem Robust emotion recognition from speech audio with temporal variations.
method Introduces multi-time-scale (MTS) convolutional layers to CNNs.
result MTS layers improve generalization, especially on smaller datasets.
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 …