|
|
Named Entity Recognition Exploiting Sub Word Information
Dobrovodský, Patrik ; Egorova, Ekaterina (oponent) ; Kesiraju, Santosh (vedoucí práce)
The aim of this thesis is the creation of a Named Entity Recognition system based on an older state-of-the-art model and studying how subword information can improve the recognition of out-of-vocabulary words. This proposed system besides English has to support two additional Indo-European languages: German and Hungarian. This work features a named entity tagger based on deep learning using pretrained and custom-trained word embeddings, sparse features, and character embeddings extracted by a Convolutional Neural Network. All these features are then processed by sequence-based (bidirectional Long Short-Term Memory) and feature-based (Conditional Random Field) approaches with the goal of achieving a F1-score similar to the work it is based on, and to compare how far present time state-of-the-art systems have evolved. The result is a system that achieves a 90.98% F1-score on the CoNLL 2003 English test dataset using pretrained word embeddings, not far behind the original work's 91.26%. For the other two languages, the model scores 89.34% on the WikiAnn German test dataset and 93.04% on the WikiAnn Hungarian test dataset with the usage of custom-trained embeddings.
|
|
|
Multi-Task Neural Networks for Speech Recognition
Egorova, Ekaterina ; Veselý, Karel (oponent) ; Karafiát, Martin (vedoucí práce)
The first part of this Master's thesis covers theoretical investigation into the principles and usage of neural networks, including their usability for the speech recognition tasks. Then it proceeds to summarize the multi-task neural networks' operating principles and some recent experiments with them. The practical part of the semester project reports changes made to a tool for neural network training which support multi-task training. Then the preparation of the settings is described, including a number of scripts written especially for this purpose. The experiments presented in the thesis explore the idea of using articulatory characteristics of phonemes as secondary tasks for multi-task training. The experiments are conducted on two different datasets of different quality and size and representing different languages - English and Vietnamese. Articulatory characteristics are occasionally combined with different secondary tasks, such as context, to see how well they function together. A comparison is made between the networks of different sizes to see how their size affects the effectiveness of multi-task training. These experiments show that multi-task training with the use of articulatory characteristics as secondary tasks can enhance training and yield better phoneme accuracy as a result. Finally, multi-task training is embedded to a speech recognition system as a feature extractor.
|
|
|
Out-of-Vocabulary Words Detection and Recovery
Egorova, Ekaterina ; Hannemann, Mirko (oponent) ; Schaaf, Thomas (oponent) ; Černocký, Jan (vedoucí práce)
The thesis explores the field of out-of-vocabulary word (OOV) processing within the task of automatic speech recognition (ASR). It defines the two separate OOV processing tasks - that of detection and recovery - and proposes success metrics for both the tasks. Different approaches to OOV detection and recovery are presented within the frameworks of hybrid and end-to-end (E2E) ASR. These approaches and compared on an open access LibriSpeech database to facilitate replicability. Hybrid approach uses modified decoding graph with phoneme substrings and utilizes full lattice representations for detection and recovery of recurrent OOVs. Recovered OOVs are added to the dictionary and the language model (LM) to improve ASR system performance. The second approach employs inner representations of a word-predicting Listen Attend and Spell architecture (LAS) E2E system to perform OOV detection task. Detection recall and precision rates improved drastically in comparison with the hybrid approach. Recur-rent OOV recovery is performed on a separate character-predicting system with the use of detected time frames and probabilistic clustering.Finally, we propose a new speller architecture with a capability of learning OOV representations together with the word predicting network (WPN) training. The speller forces word embeddings to be spelling-aware during the training and thus not only provides OOV recovery, but also improves the WPN performance.
|
|
|
Named Entity Recognition Exploiting Sub Word Information
Dobrovodský, Patrik ; Egorova, Ekaterina (oponent) ; Kesiraju, Santosh (vedoucí práce)
The aim of this thesis is the creation of a Named Entity Recognition system based on an older state-of-the-art model and studying how subword information can improve the recognition of out-of-vocabulary words. This proposed system besides English has to support two additional Indo-European languages: German and Hungarian. This work features a named entity tagger based on deep learning using pretrained and custom-trained word embeddings, sparse features, and character embeddings extracted by a Convolutional Neural Network. All these features are then processed by sequence-based (bidirectional Long Short-Term Memory) and feature-based (Conditional Random Field) approaches with the goal of achieving a F1-score similar to the work it is based on, and to compare how far present time state-of-the-art systems have evolved. The result is a system that achieves a 90.98% F1-score on the CoNLL 2003 English test dataset using pretrained word embeddings, not far behind the original work's 91.26%. For the other two languages, the model scores 89.34% on the WikiAnn German test dataset and 93.04% on the WikiAnn Hungarian test dataset with the usage of custom-trained embeddings.
|
|
|
Multi-Task Neural Networks for Speech Recognition
Egorova, Ekaterina ; Veselý, Karel (oponent) ; Karafiát, Martin (vedoucí práce)
The first part of this Master's thesis covers theoretical investigation into the principles and usage of neural networks, including their usability for the speech recognition tasks. Then it proceeds to summarize the multi-task neural networks' operating principles and some recent experiments with them. The practical part of the semester project reports changes made to a tool for neural network training which support multi-task training. Then the preparation of the settings is described, including a number of scripts written especially for this purpose. The experiments presented in the thesis explore the idea of using articulatory characteristics of phonemes as secondary tasks for multi-task training. The experiments are conducted on two different datasets of different quality and size and representing different languages - English and Vietnamese. Articulatory characteristics are occasionally combined with different secondary tasks, such as context, to see how well they function together. A comparison is made between the networks of different sizes to see how their size affects the effectiveness of multi-task training. These experiments show that multi-task training with the use of articulatory characteristics as secondary tasks can enhance training and yield better phoneme accuracy as a result. Finally, multi-task training is embedded to a speech recognition system as a feature extractor.
|
host ::
RSS.