Alternative method proposed for handling uncertainty in i-vector extraction.
problem Uncertainty in i-vector extraction for spoken language recognition.
method Proposes an alternative method to propagate uncertainty into the Gaussian back-end.
result Alternative method effectively handles uncertainty in i-vector extraction.
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…
ACI converts call center conversations into actionable data.
problem Real-time spoken language understanding for call center conversations.
method Combines speech recognition, entity and intent recognition, and a business rules engine.
result ACI converts live audio into structured events for real-time supervision and assistance.
Improved spoken English intelligibility with computer recognition and feature extraction.
problem Improving spoken English pronunciation and intelligibility.
method Automatic speech recognition using PocketSphinx alignment and feature extraction with SVM classifier probability prediction.
result SVM models achieve 82 percent agreement with human transcriptions, up from 75 percent.
System diagnoses Alzheimer's disease from spoken language using multi-modal features.
problem Early diagnosis of Alzheimer's disease from spoken language.
method Classification system based on spoken language using three approaches (N-gram, i-vector, x-vector).
result Accuracy of 83.6% on the cookie picture description task from Pitt Corpus dementia bank.
Benchmark tests spoken language models for infant language learning.
problem Understanding how infants learn language from speech.
method Developed a language-acquisition-friendly benchmark.
result Benchmarking shows models' strengths and weaknesses.
Expanding spoken language understanding to handle complex entities and intents.
problem Handling compound entities and intents in spoken language understanding.
method Introducing a domain-agnostic shallow parser that handles linguistic coordination, learning domain-independent and slot-independent features.
result The model learns to segment conjunct boundaries of various phrasal categories and improves generalization across different slot types using adversarial training.
This research investigates the Statistical Machine Translation approaches to translate speech in real time automatically. Such systems can be used in a pipeline with speech recognition and synthesis software in order to produce a real-time voice communication system between foreigners. We obtained three main data sets …
Improves domain classification across multiple locales with shared language.
problem Improves domain classification accuracy in Spoken Language Understanding across multiple locales with shared language.
method Selective multi-task learning to create a joint representation of utterances over locales with different sets of domains.
result The proposed approach outperforms other baselines models especially when classifying locale-specific domains and low-resourced domains.
Paper explores unsupervised transfer learning for SLU, improving model performance with unlabeled data.
problem Improving SLU model performance with limited labeled data.
method Uses ELMo embeddings for unsupervised pre-training and ELMo-Light for faster pre-training. Combines unsupervised and supervised transfer techniques.
result Unsupervised pre-training on unlabeled data significantly improves SLU performance, even outperforming conventional supervised transfer.
New model learns coupled representations for domains, intents, and slots.
problem Representation learning for domains, intents, and slots in spoken language understanding.
method Proposes a model that learns coupled representations by aggregating slot and intent representations based on their hierarchical relationships.
result Improved performance on contextual cross-domain reranking task.
AudioPaLM combines text and speech models to improve speech processing and translation.
problem Improving speech understanding and generation models.
method Unified multimodal architecture combining text and speech models.
result AudioPaLM significantly outperforms existing speech translation systems.
Paper presents a lightweight, unobtrusive method to protect edge device data privacy.
problem Protecting inference data privacy in IoT edge devices with limited compute power.
method A lightweight neural network at edge devices to obfuscate inference data without indicating obfuscation.
result Effectively protects inference data confidentiality while preserving backend accuracy.
Robot learns word meanings from perception-action tasks.
problem Language acquisition for robots.
method Affordance network with temporal co-occurrence of speech and actions.
result Robot forms useful word-to-meaning associations.
RUSLAN is a large Russian speech corpus for text-to-speech.
problem Lack of high-quality annotated Russian speech data for text-to-speech.
method Developed a large annotated Russian speech corpus and trained a neural network for text-to-speech synthesis.
result Synthesized speech quality evaluated with MOS scores: 4.05 for naturalness, 3.78 for intelligibility.
Neural User Simulator outperforms traditional ABUS in training dialogue systems.
problem Limited diversity and lack of natural language in ABUS.
method NUS learns user behavior from a corpus and generates natural language.
result NUS trained policies outperform ABUS in real user evaluations.
This research explores the effects of various training settings on a Polish to English Statistical Machine Translation system for spoken language. Various elements of the TED, Europarl, and OPUS parallel text corpora were used as the basis for training of language models, for development, tuning and testing of the tran…
End-to-end neural network improves QbE-STD in multilingual speech search.
problem Query by example spoken term detection in zero-resource scenarios.
method Use of multilingual bottleneck features and CNN for pattern matching, integrated into a fully neural network framework.
result CNN-based matching outperforms DTW-based matching using bottleneck features.
WEEND uses a neural network to recognize speech and assign speakers to words.
problem End-to-end neural diarization without additional ASR and orchestration.
method Multi-task learning with an auxiliary network for ASR and speaker diarization.
result WEEND outperforms turn-based diarization and can handle 5-minute audio.
NLP techniques improve drug discovery by analyzing chemical and protein text.
problem Improving drug discovery through better analysis of chemical and protein text.
method Natural language processing techniques applied to biochemical entities.
result Enhanced prediction of molecular properties and design of novel molecules.
Neural network framework for language recognition considers sequence information and improves accuracy.
problem Challenging task of automatic language identification in noisy conditions.
method Proposes a neural network framework with bidirectional LSTM and attention modeling for relevance weighting.
result Significant improvements over conventional methods in noisy conditions and multi-speaker speech.
New approach uses graphs for sign language recognition.
problem Challenges in recognizing sign language for deaf individuals.
method Spatial-Temporal Graph Convolutional Network.
result Improved sign language recognition using human skeletal movements.
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…
Improved LID for multilingual speakers using context-aware models.
problem Low accuracy for languages spoken by multilingual speakers, especially with accented speech.
method Coarser-grained acoustic model and integration with interaction context signals.
result Average 97% accuracy across all language combinations, 60% improvement in worst-case accuracy.
Learned feature representations and sub-phoneme posteriors from Deep Neural Networks (DNNs) have been used separately to produce significant performance gains for speaker and language recognition tasks. In this work we show how these gains are possible using a single DNN for both speaker and language recognition. The u…
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.
Improves E2E ASR performance on numeric sequences with additional training data and denormalization.
problem Challenges in recognizing numeric sequences out-of-vocabulary in ASR systems.
method Uses text-to-speech for additional numeric training data and a small-footprint neural network for denormalization.
result Reduction of WER by up to a factor of 8 in the longest numeric sequences.
System supports 102 languages with deep neural networks, reducing error rates and improving recognition speed.
problem Online handwriting recognition for multiple languages with high accuracy and speed.
method Deep neural network architecture, Bézier curves for input encoding, sequence recognition methods.
result Reduced error rates by 20%-40% for most languages, up to 10x faster recognition times.
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.
Study develops sign recognition system for DHH users.
problem Accessibility of voice-controlled devices for Deaf and Hard-of-Hearing users.
method Multimodal data (RGB video and skeletal data) for sign language recognition using deep learning.
result Validation on GMUASL51 dataset of 12 users and 13107 samples across 51 signs.
New framework learns phoneme metric from perception data.
problem Learn metric for phoneme similarity from data.
method Learning algorithms to derive metric from behavioral data.
result Framework outperforms previous metrics in phoneme prediction.
Paper proposes M2H-GAN to improve speech theme identification.
problem Limited ASR transcripts for speech theme identification.
method Uses M2H-GAN, a GAN-based approach, to generate TRS-like ASR transcripts.
result Improves speech theme identification performance close to human levels.
This research explores the effects of various training settings from Polish to English Statistical Machine Translation system for spoken language. Various elements of the TED parallel text corpora for the IWSLT 2013 evaluation campaign were used as the basis for training of language models, and for development, tuning …
Enhances spoken speech quality using EEG signals.
problem Improves speech clarity in noisy environments.
method Generative adversarial network (GAN), gated recurrent unit (GRU), temporal convolutional network (TCN) regression models.
result Significant improvement in speech enhancement quality compared to traditional methods.
Novel CNN array for sign language recognition using wearable IMUs.
problem Efficiently recognizing sign language from wearable IMU signals.
method Two-dimensional Convolutional Neural Network array architecture for Indian sign language recognition.
result Peak classification accuracies of 94.20% for general sentences and 95.00% for interrogative sentences achieved.
This paper maps the large-scale variation of the Spanish language by employing a corpus based on geographically tagged Twitter messages. Lexical dialects are extracted from an analysis of variants of tens of concepts. The resulting maps show linguistic variation on an unprecedented scale across the globe. We discuss th…
Paper proposes RMFN for multimodal language analysis.
problem Modeling interactions between language, visual, and acoustic modalities.
method Recurrent Multistage Fusion Network (RMFN) decomposes fusion into stages focusing on subsets of multimodal signals.
result RMFN achieves state-of-the-art performance across multimodal sentiment analysis, emotion recognition, and speaker traits recognition datasets.
Joint PLDA models two factors for better speaker verification.
problem Improving speaker verification accuracy with a nuisance condition.
method Generalizes PLDA to model two sample-dependent factors, modifies training procedure.
result Significant performance gains in multilingual speaker-verification tasks.
Standard probabilistic linear discriminant analysis (PLDA) for speaker recognition assumes that the sample's features (usually, i-vectors) are given by a sum of three terms: a term that depends on the speaker identity, a term that models the within-speaker variability and is assumed independent across samples, and a fi…
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.
This study applies neural models to automatically recognize medical entities from natural language.
problem Automated recognition of medical entities from natural language is complex and time-consuming.
method Utilizes deep neural sequence models trained on a large dataset of death certificates.
result Deep neural models can efficiently recognize medical entities from natural language.
Deep CapsNet improves sign language recognition from wearable IMUs.
problem Continuous recognition of sign language from wearable devices.
method Custom CapsNet architecture using deep capsule networks and game theory.
result Improved accuracy of 94% and 92.50% for 3 and 5 routings respectively, compared to 87.99% for CNN.
This research converts visual information into audio for users to perceive.
problem Brevity in conveying visual information through spoken language.
method Pretrained image embedding network, GAN for metric space mapping, human subject testing.
result Users can accurately classify audio sonifications of faces.
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.
ClovaCall introduces a new Korean call speech corpus for contact centers.
problem Lack of large-scale call-based speech corpora for Korean dialog scenarios.
method Development of a new large-scale Korean call-based speech corpus (ClovaCall) in a restaurant reservation domain.
result Validation of the dataset with ASR models shows its effectiveness.
Improves cross-lingual NER by projecting entities from one language to another.
problem Limited annotated corpora for named entity recognition in many languages.
method Uses machine translation twice: first for sentences, then for entities; matches based on ortho- and phonetic similarity; identifies matches using distributional statistics.
result Improves cross-lingual NER by an average of 4.1 points on 5 diverse languages.
New Polish word embeddings improve temporal expression recognition.
problem Recognizing temporal expressions in Polish text.
method Created KGR10 corpus, used BiLSTM-CRF model with custom embeddings.
result Custom embeddings enhance BiLSTM-CRF model's performance.
This study analyzes how one-layer transformers learn regular language recognition tasks.
problem Understanding how one-layer transformers solve regular language recognition tasks like even pairs and parity check.
method Theoretical analysis of training dynamics and gradient descent for a one-layer transformer.
result A one-layer transformer can solve even pairs directly but needs CoT for parity check. Training phases show rapid growth in attention layer followed by logarithmic growth in linear layer.