You are now leaving the DARPA.mil website that is under the control and management of DARPA. The appearance of hyperlinks does not constitute endorsement by DARPA of non-U.S. Government sites or the information, products, or services contained therein. Although DARPA may or may not use these sites as additional distribution channels for Department of Defense information, it does not exercise editorial control over all of the information that you may find at these locations. Such links are provided consistent with the stated purpose of this website.

After reading this message, click to continue immediately.

Go Back

/ Information Innovation Office (I2O)

Robust Automatic Transcription of Speech (RATS)

The Robust Automatic Transcription of Speech (RATS) program will create algorithms and software for performing the following tasks on potentially speech-containing signals received over communication channels that are extremely noisy and/or highly distorted: speech activity detection, language identification, speaker identification, and key word spotting.

Program Manager: Dr. David Doermann

Contact: david.doermann@darpa.mil

The content below has been generated by organizations that are partially funded by DARPA; the views and conclusions contained therein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of DARPA or the U.S. Government.

Report a problem: opencatalog@darpa.mil

Last updated: November 13, 2015

TeamTitleLink
LDC Annotation Trees: LDC's Customizable, Extensible, Scalable Annotation Infrastructure
BBN White Listing and Score Normalization for Keyword Spotting of Noisy Speech
BBN Improvements in Language Identification on the RATS Noisy Speech Corpus
(Team BBN) University of Cambridge Infinite Structured Support Vector Machines in Speech Recognition
(Team BBN) University of Cambridge An Explicit Independence Constraint for Factorised Adaptation in Speech Recognition
(Team BBN) University of Cambridge A Confidence-Based Approach for Improving Keyword Hypothesis Scores
(Team BBN) University of Cambridge TANDEM System Adaptation Using Multiple Linear Feature Transforms
(Team BBN) University of Cambridge Model-Based Approaches for Degraded Channel Modelling in Robust ASR
(Team BBN) University of Cambridge Using Sub-Word-Level Information for Confidence Estimation with Conditional Random Field Models
(Team BBN) University of Cambridge Model-Based Approaches to Adaptive Training in Reverberant Environments
(Team BBN) University of Cambridge Factor Analysis Based VTS Discriminative Adaptive Training
(Team BBN) Brno University of Technology Pairwise Discriminative Speaker Verification in the I -Vector Space
(Team BBN) Brno University of Technology, Brno University of Technology, MIT, Johns Hopkins University, University of Maryland, Aalborg University Probabilistic Linear Discriminant Analysis Of I-Vector Posterior Distributions
(Team BBN) Brno University of Technology Developing a Speaker Identification System for the DARPA RATS Project
(Team BBN) Brno University of Technology, Department of Electronics and Telecommunications, NTNU, Politecnico di Torino Regularized Subspace n-Gram Model for Phonotactic iVector Extraction
(Team BBN) Brno University of Technology, Politecnico di Torino, AGNITIO Description and Analysis of the Brno276 system for LRE2011
(Team BBN) Brno University of Technology, Universidad Politcnica de Madrid Phonotactic Language Recognition using i-Vectors and Phoneme Posteriogram Counts
(Team BBN) Brno University of Technology, MIT Patrol Team Language Identification System for DARPA RATS P1 Evaluation
(Team BBN) Brno University of Technology, University of Maryland Developing a Speech Activity Detection System for the DARPA RATS Program
(Team BBN) Brno University of Technology Speaker Vectors from Subspace Gaussian Mixture Model as Complementary Features for Language Identification
(Team BBN) JHU Long, Deep and Wide Neural Nets for Dealing with Unexpected Noise in Machine Recognition of Speech
(Team BBN) JHU Multistream Recognition of Speech: Dealing With Unknown Unknowns
(Team IBM) JHU Robust Speaker Recognition Using Spectro-Temporal Autoregressive Models
(Team IBM) LSCP, ENS/EHESS/CNRS, INRIA/ENS/CNRS, Johns Hopkins University Evaluating Speech Features with the Minimal-Pair ABX task: Analysis of the Classical MFC/PLP Pipeline
(Team BBN) JHU Multi-Stream Recognition of Noisy Speech with Performance Monitoring
(Team BBN) JHU Text-to-Speech Inspired Duration Modeling for Improved Whole-Word Acoustic Models
(Team BBN) JHU Weak Top-Down Constraints For Unsupervised Acoustic Model Training
(Team BBN) JHU Deep Neural Network Features and Semi-Supervised Training for Low Resource Speech Recognition
(Team BBN) JHU Effect Of Filter Bandwidth and Spectral Sampling Rate of Analysis Filterbank on Automatic Phoneme Recognition
(Team BBN) JHU Frequency Offset Correction in Speech Without Detecting Pitch
(Team BBN) JHU A Summary Of The 2012 JHU CLSP Workshop on Zero Resource Speech Technologies and Models of Early Language Acquisition
(Team BBN) JHU Mean Temporal Distance: Predicting ASR Error from Temporal Properties of Speech Signal
(Team BBN) JHU Filter-Bank Optimization for Frequency Domain Linear Prediction
(Team BBN) JHU Estimating Classifier Performance in Unknown Noise
(Team BBN) JHU Data-Driven Posterior Features for Low Resource Speech Recognition Applications
(Team BBN) JHU Phone Recognition in Critical Bands Using Sub-Band Temporal Modulations
(Team BBN) JHU MAP Estimation of Whole-Word Acoustic Models with Dictionary Priors
(Team BBN) JHU Inverting the Point Process Model for Fast Phonetic Keyword Search
(Team BBN) JHU Analysis of Temporal Resolution in Frequency Domain Linear Prediction
(Team BBN) JHU Multilingual MLP Features For Low-Resource LVCSR Systems
(Team IBM) University of Maryland Automatic Intelligibility Assessment of Pathologic Speech in Head and Neck Cancer Based on Auditory-Inspired Spectro-Temporal Modulations
(Team IBM) MIT Bayesian Distance Metric Learning on i-Vector for Speaker Verification
(Team IBM) MIT Gaussian Mixture Model Weight Supervector Decomposition and Adaptation
IBM The IBM Speech Activity Detection System for the DARPA RATS Program
IBM Using Polynomial Kernel Support Vector Machines for Speaker Verification, in IEEE Letters
IBM The IBM RATS Phase II Speaker Recognition System: Overview and Analysis, in Proc. of Interspeech
IBM Unifying PLDA and Polynomial Kernel SVMs, in Proc. of IEEE ICASSP
IBM Frame-Based Phonotactic Language Identification, SLT
IBM Noisy Channel Adaptation in Language Identification, SLT
IBM On the Use of Non-Linear Polynomial Kernel SVMs in Language Recognition, Interspeech
IBM Speech Activity Detection for Noisy Data Using Adaptation Techniques, Interspeech
IBM Unsupervised Channel Adaptation for Language Identification Using Co-Training, ICASSP
IBM Trap Language Identification System for RATS Phase II Evaluation, Interspeech
IBM Enhancing Frequency Shifted Speech Signals in Single Side Band Communication, IEEE Signal Processing Letters
(Team IBM) University of Southern California, Technical University of Crete Speaker Verification using Simplified and Supervised I-Vector Modeling
(Team IBM) University of Southern California Multi-band Long-term Signal Variability Features for Robust Voice Activity Detection
(Team IBM) University of Southern California TRAP Language Identification System for RATS Phase II Evaluation
(Team IBM) University of Southern California A Robust Frontend for VAD: Exploiting Contextual, Discriminative and Spectral Cues of Human Voice
(Team IBM) University of Southern California Simplified Supervised I-vector Modeling and Sparse Representation with Application to Robust Language Recognition
(Team IBM) University of Southern California A Study on the Effect of Prosodic Emphasis Transfer on Overall Speech Translation Quality
(Team IBM) University of Southern California Spectro-Temporal Directional Derivative Features for Automatic Speech Recognition
LDC LDC Forced Aligner
LDC The RATS Radio Traffic Collection System
(Team SRI) University of California - Los Angeles A Novel Approach to Soft-Mask Estimation and Log-Spectral Enhancement for Robust Speech Recognition
(Team SRI) University of California - Los Angeles A Pitch-Based Spectral Enhancement Technique for Robust Speech Processing
SRI Adaptive Gaussian Backend for Robust Language Identification
SRI All For One: Feature Combination For Highly Channel-Degraded Speech Activity Detection
(Team SRI) Incheon National University, University of Texas - Dallas An Advanced Feature Compensation Method Employing Acoustic Model with Phonetically Constrained Structure
SRI Bilinear Factor Analysis for iVector Based Speaker Verification
(Team SRI) USC Berkeley, ICSI Confidence-Based Scoring: A Useful Diagnostic Tool for Detection Tasks
SRI Damped Oscillator Cepstral Coefficients for Robust Speech Recognition
SRI Discriminatively Trained Phoneme Confusion Model for Keyword Spotting
(Team SRI) USC Berkeley, ICSI Easy Does It: Robust Spectro-Temporal Many-Stream ASR Without Fine Tuning Streams
(Team SRI) University of Texas - Dallas Factor Analysis of Acoustic Features Using a Mixture of Probabilistic Principal Component Analyzers for Robust Speaker Verification
(Team SRI) University of Texas - Dallas Feature Compensation Employing Online GMM Adaptation for Speech Recognition in Unknown Severely Adverse Environments
(Team SRI) University of Texas - Dallas Gaussian Map based Acoustic Model Adaptation Using Untranscribed Data for Speech Recognition in Severely Adverse Environments
(Team SRI) CMU Histogram-Based Subband Power Warping and Spectral Averaging for Robust Speech Recognition Under Matched and Multistyle Training
(Team SRI) University of Texas - Dallas Impact of Noise Reduction and Spectrum Estimation on Noise Robust Speaker Identification
SRI Improving Language Identification Robustness to Highly Channel-Degraded Speech through Multiple System Fusion
(Team SRI) USC Berkeley, ICSI Informative Spectro-Temporal Bottleneck Features for Noise-Robust Speech Recognition
(Team SRI) USC Berkeley, ICSI Longer Features: They Do a Speech Detector Good
(Team SRI) USC Berkeley, ICSI Low Complexity Spectral Imputation for Noise Robust Speech Recognition
(Team SRI) University of Texas - Dallas Mean Hilbert Envelope Coefficients (MHEC) for Robust Speaker Recognition
(Team SRI) University of California - Los Angeles Modulation Features for Noise Robust Speaker Identification
(Team SRI) University of California - Los Angeles Multi-Band Summary Correlogram-Based Pitch Detection for Noisy Speech
(Team SRI) Columbia University, International Computer Science Institute Noise Robust Pitch Tracking by Subband Autocorrelation Classification
SRI Normalized Amplitude Modulation Features for Large Vocabulary Noise-Robust Speech Recognition
(Team SRI) University of Texas - Dallas Phoneme Class Based Adaptation for Mismatch Acoustic Modeling of Distant Noisy Speech
(Team SRI) CMU, Microsoft Power-Normalized Cepstral Coefficients (PNCC) for Robust Speech Recognition
(Team SRI) University of Texas - Dallas Robust Front-End Processing For Speaker Identification Over Extremely Degraded Communication Channels
(Team SRI) USC Berkeley, ICSI Spectro-Temporal Features for Robust Speech Recognition Using Power-Law Nonlinearity and Power-Bias Subtraction
(Team SRI) USC Berkeley, ICSI Spectro-Temporal Gabor Features for Speaker Recognition
(Team SRI) USC Berkeley, ICSI Speech Activity Detection: An Economics Approach
(Team SRI) USC Berkeley, ICSI Strategies for High Accuracy Keyword Detection in Noisy Channels
(Team SRI) University of Texas - Dallas Unsupervised Speech Activity Detection using Voicing Measures and Perceptual Spectral Flux
Efficient Lattice Rescoring Using Recurrent Neural Network Language Models
On the Use of i-Vector Posterior Distributions in Probabilistic Linear Discriminant Analysis
Non-negative Factor Analysis of Gaussian Mixture Model Weight Adaption for Language and Dialect Recognition
GMM Weights Adaptation Based on Subspace Approaches for Speaker Verification
Unscented Transform For i-Vector-based Noisy Speaker Recognition