All

What are you looking for?

All
Projects
Results
Organizations

Quick search

  • Projects supported by TA ČR
  • Excellent projects
  • Projects with the highest public support
  • Current projects

Smart search

  • That is how I find a specific +word
  • That is how I leave the -word out of the results
  • “That is how I can find the whole phrase”

Filters

49 120 (0,106s)

Result

Speaker adaptation in the Czech large vocabulary continuous speech recognition system

large vocabulary continuous speech recognition system. Speaker independent systems are usually trained by the speech data of many speakers but achieve a lower recognition rate than speaker dependent systems. A nee...

JD - Využití počítačů, robotika a její aplikace

  • 2001
  • D
Result

Speaker Adaptation of Language Models for Automatic Dialog Act Segmentation of Meetings

understanding. In this paper, we explore speaker adaptation of hidden event language models (LMs) for DA segmentation using the ICSI Meeting Corpus. Speaker adaptation trained on only the data from indivi...

JD - Využití počítačů, robotika a její aplikace

  • 2007
  • D
Result

Speaker Adaptation of Language Models for Automatic Dialog Act Segmentation of Meetings

understanding. In this paper, we explore speaker adaptation of hidden event language models (LMs) for DA segmentation using the ICSI Meeting Corpus. Speaker adaptation trained on only the data from indivi...

JD - Využití počítačů, robotika a její aplikace

  • 2007
  • Jx
Result

On Speaker Adaptive Training of Artificial Neural Networks

of multilayer perceptron neural networks (MLP ANN) by means of adopting Speaker Adaptive Training. The use of the MLP ANN, usually in combination with the TRAPS parametrization for GMM-HMM and other. In the first SAT exper...

JD - Využití počítačů, robotika a její aplikace

  • 2010
  • D
Result

Optimization of Speaker-aware Multichannel Speech Extraction with ASR Criterion

, we inform the neural network about the target speaker using information extracted from an adaptation utterance, enabling the network to track the target speaker. While in the previous work, this method was used to separat...

Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

  • 2018
  • D
  • Link
Result

Online Speaker Adaptation of an Acoustic Model Using Face Recognition

We have proposed and evaluated a novel approach for online speaker adaptation speaker independent acoustic models, gender dependent models and models of particular speakers. In the experiment, the speaker ...

JD - Využití počítačů, robotika a její aplikace

  • 2013
  • D
  • Link
Result

Training of Speaker-Clustered Acoustic Models for Use in Real-Time Recognizers

The paper deals with training of speaker-clustered acoustic models. Various training techniques - Maximum Likelihood, Discriminative Training and two adaptation based on the MAP and Discriminative MAP were...

JD - Využití počítačů, robotika a její aplikace

  • 2009
  • D
Result

Speaker Verification Using End-To-End Adversarial Language Adaptation

In this paper we investigate the use of adversarial domain adaptation for addressing the problem of language mismatch between speaker recognition corpora. In the context of speaker verification, adversarial domain adapt...

Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

  • 2019
  • D
  • Link
Result

Parameter-Efficient Transfer Learning of Pre-Trained Transformer Models for Speaker Verification Using Adapters

learnable parameters for adapting to speaker verification tasks. Specifically, during the fine-tuning process, the pre-trained models are frozen, and only lightweightRecently, the pre-trained Transformer models ha...

Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

  • 2023
  • D
  • Link
Result

Speaker adaptation for Wav2vec2 based dysarthric ASR

of training data and heavy mismatch in speaker characteristics. Recent ASR systems have the recognition performance. Speaker adaptation using fMLLR and xvectors have provided major gains for dysarthric speech with...

Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

  • 2022
  • D
  • Link
  • 1 - 10 out of 49 120