Automatic transcription of the Polish newsreel

2019 ◽  
Vol 55 (2) ◽  
pp. 183-209
Author(s):  
Danijel Koržinek ◽  
Krzysztof Wołk ◽  
Łukasz Brocki ◽  
Krzysztof Marasek
Keyword(s):  
Speech Recognition ◽  
Recognition System ◽  
Language Models ◽  
Late 20Th Century ◽  
Word Error Rate ◽  
Audio Quality ◽  
Transcription System ◽  
Automatic Transcription ◽  
Using Data ◽  
Optimal Adaptation

Abstract This paper describes an automatic transcription system for the Polish Newsreel, which is a collection of mid to late 20th century news segments presented in audio and video form. They are characterized by their use of archaic language and poor audio quality, which makes them a demanding problem for speech recognition systems. Acoustic and language models had to be retrained using data from in-domain corpora. During the adaptation of the models, experiments were carried out to select optimal adaptation parameters. The experiments showed that the adaptation of the speech recognition system to a narrow and clearly defined domain significantly increases its efficiency. The final word error rate obtained for this domain was 10.97%.

scholarly journals Hierarchical Phoneme Classification for Improved Speech Recognition

2021 ◽  
Vol 11 (1) ◽  
pp. 428
Author(s):  
Donghoon Oh ◽  
Jeong-Sik Park ◽  
Ji-Hwan Kim ◽  
Gil-Jin Jang
Keyword(s):  
Speech Recognition ◽  
Language Processing ◽  
Confusion Matrix ◽  
Critical Factor ◽  
Recognition System ◽  
Classification Performance ◽  
Language Models ◽  
Successful Implementation ◽  
Phoneme Classification ◽  
Improved Performance

Speech recognition consists of converting input sound into a sequence of phonemes, then finding text for the input using language models. Therefore, phoneme classification performance is a critical factor for the successful implementation of a speech recognition system. However, correctly distinguishing phonemes with similar characteristics is still a challenging problem even for state-of-the-art classification methods, and the classification errors are hard to be recovered in the subsequent language processing steps. This paper proposes a hierarchical phoneme clustering method to exploit more suitable recognition models to different phonemes. The phonemes of the TIMIT database are carefully analyzed using a confusion matrix from a baseline speech recognition model. Using automatic phoneme clustering results, a set of phoneme classification models optimized for the generated phoneme groups is constructed and integrated into a hierarchical phoneme classification method. According to the results of a number of phoneme classification experiments, the proposed hierarchical phoneme group models improved performance over the baseline by 3%, 2.1%, 6.0%, and 2.2% for fricative, affricate, stop, and nasal sounds, respectively. The average accuracy was 69.5% and 71.7% for the baseline and proposed hierarchical models, showing a 2.2% overall improvement.

USING DATA-DRIVEN SUBWORD UNITS IN LANGUAGE MODEL OF HIGHLY INFLECTIVE SLOVENIAN LANGUAGE

2009 ◽  
Vol 23 (02) ◽  
pp. 287-312 ◽  
Author(s):  
MIRJAM SEPESY MAUČEC ◽  
TOMAŽ ROTOVNIK ◽  
ZDRAVKO KAČIČ ◽  
JANEZ BREST
Keyword(s):  
Speech Recognition ◽  
Language Model ◽  
Recognition System ◽  
Language Models ◽  
Data Driven ◽  
Inflectional Morphology ◽  
Minimum Entropy ◽  
Test Set ◽  
Training Corpus ◽  
Speech Database

This paper presents the results of a study on modeling the highly inflective Slovenian language. We focus on creating a language model for a large vocabulary speech recognition system. A new data-driven method is proposed for the induction of inflectional morphology into language modeling. The research focus is on data sparsity, which results from the complex morphology of the language. The idea of using subword units is examined. An attempt is made to figure out the segmentation of words into two subword units: stems and endings. No prior knowledge of the language is used. The subword units should fit into the frameworks of the probabilistic language models. A morphologically correct decomposition of words is not being sought, but searching for a decomposition which yields the minimum entropy of the training corpus. This entropy is approximated by using N-gram models. Despite some seemingly over-simplified assumption, the subword models improve the applicability of the language models for a sparse training corpus. The experiments were performed using the VEČER newswire text corpus as a training corpus. The test set was taken from the SNABI speech database, because the final models were evaluated in speech recognition experiments on SNABI speech database. Two different subword-based models are proposed and examined experimentally. The experiments demonstrate that subword-based models, which considerably reduce OOV rate, improve speech recognition WER when compared with standard word-based models, even though they increase test set perplexity. Subword-based models with improved perplexity, but which reduce the OOV rate much less than the previous ones, do not improve speech recognition results.

scholarly journals Augmented Latent Features of Deep Neural Network-Based Automatic Speech Recognition for Motor-Driven Robots

2020 ◽  
Vol 10 (13) ◽  
pp. 4602
Author(s):  
Moa Lee ◽  
Joon-Hyuk Chang
Keyword(s):  
Speech Recognition ◽  
Recognition System ◽  
Recognition Algorithm ◽  
Internal Layers ◽  
Acoustic Features ◽  
Word Error Rate ◽  
State Information ◽  
Intelligent Robots ◽  
State Dependent ◽  
Latent Features

Speech recognition for intelligent robots seems to suffer from performance degradation due to ego-noise. The ego-noise is caused by the motors, fans, and mechanical parts inside the intelligent robots especially when the robot moves or shakes its body. To overcome the problems caused by the ego-noise, we propose a robust speech recognition algorithm that uses motor-state information of the robot as an auxiliary feature. For this, we use two deep neural networks (DNN) in this paper. Firstly, we design the latent features using a bottleneck layer, one of the internal layers having a smaller number of hidden units relative to the other layers, to represent whether the motor is operating or not. The latent features maximizing the representation of the motor-state information are generated by taking the motor data and acoustic features as the input of the first DNN. Secondly, once the motor-state dependent latent features are designed at the first DNN, the second DNN, accounting for acoustic modeling, receives the latent features as the input along with the acoustic features. We evaluated the proposed system on LibriSpeech database. The proposed network enables efficient compression of the acoustic and motor-state information, and the resulting word error rate (WER) are superior to that of a conventional speech recognition system.

Sign in / Sign up

Export Citation Format

Share Document