Add to ORKGSpeech recognition systems are usually speaker-inde-pendent, but they are not as good as speaker-dependent systems for specific speakers. An initial speaker-indepen-dent system can be adapted to improve recognition accu-racy by transforming it into a speaker-dependent system. In this work, a new general acoustic model adaptation technology is presented, using the MLLR algorithm it-eratively in a supervised manner. Experiments have been performed on the TT2 Spanish speech corpus. The initial acoustic models were trained from the Albayzin speech database. Their results, which were obtained for 10 speak-ers, show an improvement in speech recognition accuracy. 1
EUROSPEECH2001: the 7th European Conference on Speech Communication and Technology, September 3-7, ...
In this paper, we propose a novel speaker adaptation technique, regularized-MLLR, for Computer Assis...
A novel speech feature generation-based acoustic model training method for robust speaker-independen...
In this paper an effective technique for speaker adaptation on the feature domain is presented. This...
The performance of the speech recognition systems to translate voice to text is still an issue in la...
To recognize non-native speech, larger acoustic/linguistic distor-tions must be handled adequately i...
The speaker-dependent HMM-based recognizers gives lower word error rates in comparison with the corr...
Inter-speaker variation can be coped rather well in speech recognition by speaker adaptation techniq...
This paper presents a technical speaker adaptation method called WMLLR, which is based on maximum li...
A robust ASR system needs to perform well in different environment and with different speakers. For ...
Multilingual Automatic Speech Recognition (ASR) systems are of great interest in multilingual enviro...
The goal of this thesis is to find new and efficient features for speaker recognition. We are mostly...
This paper considers the problem of speaker adaptation of acous-tic models in speech recognition. We...
Abstract—We present a new modeling approach for speaker recognition that uses the maximum-likelihood...
Though speaker adaptation has long been an importing topic in automatic speech recognition, the brea...
EUROSPEECH2001: the 7th European Conference on Speech Communication and Technology, September 3-7, ...
In this paper, we propose a novel speaker adaptation technique, regularized-MLLR, for Computer Assis...
A novel speech feature generation-based acoustic model training method for robust speaker-independen...
In this paper an effective technique for speaker adaptation on the feature domain is presented. This...
The performance of the speech recognition systems to translate voice to text is still an issue in la...
To recognize non-native speech, larger acoustic/linguistic distor-tions must be handled adequately i...
The speaker-dependent HMM-based recognizers gives lower word error rates in comparison with the corr...
Inter-speaker variation can be coped rather well in speech recognition by speaker adaptation techniq...
This paper presents a technical speaker adaptation method called WMLLR, which is based on maximum li...
A robust ASR system needs to perform well in different environment and with different speakers. For ...
Multilingual Automatic Speech Recognition (ASR) systems are of great interest in multilingual enviro...
The goal of this thesis is to find new and efficient features for speaker recognition. We are mostly...
This paper considers the problem of speaker adaptation of acous-tic models in speech recognition. We...
Abstract—We present a new modeling approach for speaker recognition that uses the maximum-likelihood...
Though speaker adaptation has long been an importing topic in automatic speech recognition, the brea...
EUROSPEECH2001: the 7th European Conference on Speech Communication and Technology, September 3-7, ...
In this paper, we propose a novel speaker adaptation technique, regularized-MLLR, for Computer Assis...
A novel speech feature generation-based acoustic model training method for robust speaker-independen...