Publications related to The Kaldi Speech Recognition Toolkit

,

The i-vector and Joint Factor Analysis (JFA) systems for text- dependent speaker verification use sufficient statistics computed from a speech utterance to estimate speaker models. These statis- tics average the acoustic information over the utterance ther ...

IEEE2016

DEEP NEURAL NETWORK BASED POSTERIORS FOR TEXT-DEPENDENT SPEAKER VERIFICATION

Petr Motlicek, Subhadeep Dey

The i-vector and Joint Factor Analysis (JFA) systems for text- dependent speaker verification use sufficient statistics computed from a speech utterance to estimate speaker models. These statis- tics average the acoustic information over the utterance ther ...

Idiap2016

Automatic social role recognition and its application in structuring multiparty interactions

Ashtosh Sapru

Automatic processing of multiparty interactions is a research domain with important applications in content browsing, summarization and information retrieval. In recent years, several works have been devoted to find regular patterns which speakers exhibit ...

EPFL2015

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

Petr Motlicek, Philip Neil Garner

This paper investigates employment of Subspace Gaussian Mixture Models (SGMMs) for acoustic model adaptation towards different accents for English speech recognition. The SGMMs comprise globally-shared and state-specific parameters which can efficiently be ...

2013

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

Petr Motlicek, Philip Neil Garner

This paper investigates employment of Subspace Gaussian Mixture Models (SGMMs) for acoustic model adaptation towards different accents for English speech recognition. The SGMMs comprise globally-shared and state-specific parameters which can efficiently be ...

Idiap2013

The Kaldi Speech Recognition Toolkit

Petr Motlicek

We describe the design of Kaldi, a free, open-source toolkit for speech recognition research. Kaldi provides a speech recognition system based on finite-state automata (using the freely available OpenFst), together with detailed documentation and a compreh ...

Idiap2012

Conversion of Recurrent Neural Network Language Models to Weighted Finite State Transducers for Automatic Speech Recognition

Petr Motlicek

Recurrent neural network language models (RNNLMs) have recently shown to outperform the venerable n-gram language models (LMs). However, in automatic speech recognition (ASR), RNNLMs were not yet used to directly decode a speech signal. Instead, RNNLMs are ...

2012

Recurrent neural network language models (RNNLMs) have recently shown to outperform the venerable n-gram language models (LMs). However, in automatic speech recognition (ASR), RNNLMs were not yet used to directly decode a speech signal. Instead, RNNLMs are ...

Idiap2012

Boosting Localized Features for Speaker and Speech Recognition

Anindya Roy

In this thesis, we propose a novel approach for speaker and speech recognition involving localized, binary, data-driven features. The proposed approach is largely inspired by similar localized approaches in the computer vision domain. The success of these ...

EPFL2011

Boosting Localized Features for Speaker and Speech Recognition

Anindya Roy

In this thesis, we propose a novel approach for speaker and speech recognition involving localized, binary, data-driven features. The proposed approach is largely inspired by similar localized approaches in the computer vision domain. The success of these ...

Ecole Polytechnique Federale de Lausanne (EPFL)2011

The Kaldi Speech Recognition Toolkit

Graph Chatbot

Chat with Graph Search

DEEP NEURAL NETWORK BASED POSTERIORS FOR TEXT-DEPENDENT SPEAKER VERIFICATION

Automatic social role recognition and its application in structuring multiparty interactions

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

The Kaldi Speech Recognition Toolkit

Conversion of Recurrent Neural Network Language Models to Weighted Finite State Transducers for Automatic Speech Recognition

Boosting Localized Features for Speaker and Speech Recognition

Boosting Localized Features for Speaker and Speech Recognition

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

Automatic social role recognition and its application in structuring multiparty interactions

DEEP NEURAL NETWORK BASED POSTERIORS FOR TEXT-DEPENDENT SPEAKER VERIFICATION

ACCENT ADAPTATION USING SUBSPACE GAUSSIAN MIXTURE MODELS

Conversion of Recurrent Neural Network Language Models to Weighted Finite State Transducers for Automatic Speech Recognition

The Kaldi Speech Recognition Toolkit

Boosting Localized Features for Speaker and Speech Recognition

Boosting Localized Features for Speaker and Speech Recognition