Krzysztof Szklanny
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Krzysztof Szklanny has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Alicja Wieczorkowska
Polish-Japanese Academy of Information Technology
6 shared publications - Marcin Wichrowski
Polish-Japanese Institute of Information Technology
6 shared publications - Anna Tylki-Szymańska
Department of Pediatrics, Nutrition and Metabolic Diseases, The Children’s Memorial Health Institute, 04-730 Warsaw, Poland
4 shared publications Danijel Koržinek
Polish-Japanese Academy of Information Technology Warsaw, Poland
3 shared publications- Piotr Wrzeciono
Institute of Information Technology, Warsaw University of Life Sciences, 02-776 Warsaw, Poland
3 shared publications - Ryszard Gubrynowicz
Multimedia Department, Polish-Japanese Academy of Information Technology, Warsaw, Poland
3 shared publications
Latest Publications
- 3 March 2025
- Published by Springer Nature
- p. 125-143
A Solution for Developing Corpora for Polish Speech Enhancement in Complex Acoustic Environments
- 3 March 2025
- Published by Springer Nature
- p. 125-143
A Solution for Developing Corpora for Polish Speech Enhancement in Complex Acoustic Environments
- Open Access
- 16 January 2025
- Published by Polish Academy of Sciences Chancellery in Archives of Acoustics
Method for Vocal Fold Paralysis Detection Based on Perceptual and Acoustic Assessment
This study is aimed to evaluate a method for distinguishing between healthy and pathological voices. The evaluation was carried out using several acoustic parameters including COVAREP (collaborative voice analysis repository for speech technologies), the auditory-perceptual RBH (roughness, breathiness, hoarseness) scale, and AVQI (acoustic voice quality index). Finally, a classifier is trained using machine learning algorithms from the WEKA (Waikato Environment for Knowledge Analysis) platform. The study group comprised 75 voice recordings of individuals affected by vocal fold paralysis. The control group consisted of 49 voice recordings of healthy individuals. The results indicate that the voice quality of the study group is significantly different than the voice quality of the control group. Acoustic parameters implemented in COVAREP and the RBH scale have proven to be reliable methods assessing voice quality. In addition, data classification achieved over 90 % accuracy for every classifier.
- 16 January 2025
- Published by Polish Academy of Sciences Chancellery in Archives of Acoustics
Open AccessMethod for Vocal Fold Paralysis Detection Based on Perceptual and Acoustic Assessment
This study is aimed to evaluate a method for distinguishing between healthy and pathological voices. The evaluation was carried out using several acoustic parameters including COVAREP (collaborative voice analysis repository for speech technologies), the auditory-perceptual RBH (roughness, breathiness, hoarseness) scale, and AVQI (acoustic voice quality index). Finally, a classifier is trained using machine learning algorithms from the WEKA (Waikato Environment for Knowledge Analysis) platform. The study group comprised 75 voice recordings of individuals affected by vocal fold paralysis. The control group consisted of 49 voice recordings of healthy individuals. The results indicate that the voice quality of the study group is significantly different than the voice quality of the control group. Acoustic parameters implemented in COVAREP and the RBH scale have proven to be reliable methods assessing voice quality. In addition, data classification achieved over 90 % accuracy for every classifier.
- 31 December 2024
- Published by Krajowa Szkola Sadownictwa i Prokuratury
- p. 99-113
Perspektywy zastosowania sztucznej inteligencji i potencjał technik multimedialnych w kształceniu kadr wymiaru sprawiedliwości
This article described practical applications of multimedia techniques based on artificial intelligence (AI) solutions. Their scope includes generative technologies, i.e. creating text, images, and 360-degree panoramas. IT terms such as corpus language model are defined, which will facilitate understanding of how AI algorithms work. The article shows how these technologies can be used in judicial training. The European Judicial Training Network organized the first-ever training using VR (Virtual Reality in Judicial Training) technology. The training involved creating a presentation using 360° panoramas with artificial intelligence. The article also describes an example of cooperation between two universities – the National School of Judiciary and Public Prosecution and the Polish-Japanese Academy of Information Technology – in which students of the Faculty of New Media Arts prepared virtual tours. The prepared materials can be used in the education of judicial personnel. The advantages and disadvantages of existing solutions are discussed using specific examples. The article presents one of the tools developed as part of the CLARIN project. Common Language Resources & Technology Infrastructure) is a pan-European scientific infrastructure that allows researchers in the humanities and social sciences to work comfortably with large collections of texts. Aspects of practical applications with concrete examples are presented to implement them into everyday work in judicial education.
- 31 December 2024
- Published by Krajowa Szkola Sadownictwa i Prokuratury
- p. 99-113
Perspektywy zastosowania sztucznej inteligencji i potencjał technik multimedialnych w kształceniu kadr wymiaru sprawiedliwości
This article described practical applications of multimedia techniques based on artificial intelligence (AI) solutions. Their scope includes generative technologies, i.e. creating text, images, and 360-degree panoramas. IT terms such as corpus language model are defined, which will facilitate understanding of how AI algorithms work. The article shows how these technologies can be used in judicial training. The European Judicial Training Network organized the first-ever training using VR (Virtual Reality in Judicial Training) technology. The training involved creating a presentation using 360° panoramas with artificial intelligence. The article also describes an example of cooperation between two universities – the National School of Judiciary and Public Prosecution and the Polish-Japanese Academy of Information Technology – in which students of the Faculty of New Media Arts prepared virtual tours. The prepared materials can be used in the education of judicial personnel. The advantages and disadvantages of existing solutions are discussed using specific examples. The article presents one of the tools developed as part of the CLARIN project. Common Language Resources & Technology Infrastructure) is a pan-European scientific infrastructure that allows researchers in the humanities and social sciences to work comfortably with large collections of texts. Aspects of practical applications with concrete examples are presented to implement them into everyday work in judicial education.
- Open Access
- 9 September 2024
- Published by International Conference on Information Systems Development
Developing a Corpus for Polish Speech Enhancement by Reducing Noise, Reverberation, and Disruptions
This paper presents a solution for generating corpora of simulated Polish speech recordings in complex acoustic environments. The proposed method introduces an additional layer of unpredictable sound events, in addition to the acoustic scene noise and reverberation, making the solution unique. We generated a corpus comprising over 277 hours of training examples and over 5.5 hours for testing purposes using publicly available data sources. Next, we trained the Conv-TasNet network on the generated data to enhance single speech and separate two speakers from complex noise. The results of the experiments indicated the potential of the generated corpora for solving these tasks. Researchers can use publicly available codes to create their corpora tailored to the Polish language and solve various speech-related tasks.
- 9 September 2024
- Published by International Conference on Information Systems Development
Open AccessDeveloping a Corpus for Polish Speech Enhancement by Reducing Noise, Reverberation, and Disruptions
This paper presents a solution for generating corpora of simulated Polish speech recordings in complex acoustic environments. The proposed method introduces an additional layer of unpredictable sound events, in addition to the acoustic scene noise and reverberation, making the solution unique. We generated a corpus comprising over 277 hours of training examples and over 5.5 hours for testing purposes using publicly available data sources. Next, we trained the Conv-TasNet network on the generated data to enhance single speech and separate two speakers from complex noise. The results of the experiments indicated the potential of the generated corpora for solving these tasks. Researchers can use publicly available codes to create their corpora tailored to the Polish language and solve various speech-related tasks.
- Open Access
Implementing a Statistical Parametric Speech Synthesis System for a Patient with Laryngeal Cancer
Total laryngectomy, i.e., the surgical removal of the larynx, has a profound influence on a patient’s quality of life. The procedure results in a loss of natural voice, which in effect constitutes a significant socio-psychological problem for the patient. The main aim of the study was to develop a statistical parametric speech synthesis system for a patient with laryngeal cancer, on the basis of the patient’s speech samples recorded shortly before the surgery and to check if it was possible to generate speech quality close to that of the original recordings. The recording made use of a representative corpus of the Polish language, consisting of 2150 sentences. The recorded voice proved to indicate dysphonia, which was confirmed by the auditory-perceptual RBH scale (roughness, breathiness, hoarseness) and by acoustical analysis using AVQI (The Acoustic Voice Quality Index). The speech synthesis model was trained using the Merlin repository. Twenty-five experts participated in the MUSHRA listening tests, rating the synthetic voice at 69.4 in terms of the professional voice-over talent recording, on a 0–100 scale, which is a very good result. The authors compared the quality of the synthetic voice to another model of synthetic speech trained with the same corpus, but where a voice-over talent provided the recorded speech samples. The same experts rated the voice at 63.63, which means the patient’s synthetic voice with laryngeal cancer obtained a higher score than that of the talent-voice recordings. As such, the method enabled for the creation of a statistical parametric speech synthesizer for patients awaiting total laryngectomy. As a result, the solution would improve the quality of life as well as better mental wellbeing of the patient.
Open AccessImplementing a Statistical Parametric Speech Synthesis System for a Patient with Laryngeal Cancer
Total laryngectomy, i.e., the surgical removal of the larynx, has a profound influence on a patient’s quality of life. The procedure results in a loss of natural voice, which in effect constitutes a significant socio-psychological problem for the patient. The main aim of the study was to develop a statistical parametric speech synthesis system for a patient with laryngeal cancer, on the basis of the patient’s speech samples recorded shortly before the surgery and to check if it was possible to generate speech quality close to that of the original recordings. The recording made use of a representative corpus of the Polish language, consisting of 2150 sentences. The recorded voice proved to indicate dysphonia, which was confirmed by the auditory-perceptual RBH scale (roughness, breathiness, hoarseness) and by acoustical analysis using AVQI (The Acoustic Voice Quality Index). The speech synthesis model was trained using the Merlin repository. Twenty-five experts participated in the MUSHRA listening tests, rating the synthetic voice at 69.4 in terms of the professional voice-over talent recording, on a 0–100 scale, which is a very good result. The authors compared the quality of the synthetic voice to another model of synthetic speech trained with the same corpus, but where a voice-over talent provided the recorded speech samples. The same experts rated the voice at 63.63, which means the patient’s synthetic voice with laryngeal cancer obtained a higher score than that of the talent-voice recordings. As such, the method enabled for the creation of a statistical parametric speech synthesizer for patients awaiting total laryngectomy. As a result, the solution would improve the quality of life as well as better mental wellbeing of the patient.
- Open Access
Prototyping Mobile Storytelling Applications for People with Aphasia
Aphasia is a partial or total loss of the ability to articulate ideas or comprehend spoken language, resulting from brain damage, in a person whose language skills were previously normal. Our goal was to find out how a storytelling app can help people with aphasia to communicate and share daily experiences. For this purpose, the Aphasia Create app was created for tablets, along with Aphastory for the Google Glass device. These applications facilitate social participation and enhance quality of life by using visual storytelling forms composed of photos, drawings, icons, etc., that can be saved and shared. We performed usability tests (supervised by a neuropsychologist) on six participants with aphasia who were able to communicate. Our work contributes (1) evidence that the functions implemented in the Aphasia Create tablet app suit the needs of target users, but older people are often not familiar with tactile devices, (2) reports that the Google Glass device may be problematic for persons with right-hand paresis, and (3) a characterization of the design guidelines for apps for aphasics. Both applications can be used to work with people with aphasia, and can be further developed. Aphasic centers, in which the apps were presented, expressed interest in using them to work with patients. The Aphasia Create app won the Enactus Poland National Competition in 2015.
Open AccessPrototyping Mobile Storytelling Applications for People with Aphasia
Aphasia is a partial or total loss of the ability to articulate ideas or comprehend spoken language, resulting from brain damage, in a person whose language skills were previously normal. Our goal was to find out how a storytelling app can help people with aphasia to communicate and share daily experiences. For this purpose, the Aphasia Create app was created for tablets, along with Aphastory for the Google Glass device. These applications facilitate social participation and enhance quality of life by using visual storytelling forms composed of photos, drawings, icons, etc., that can be saved and shared. We performed usability tests (supervised by a neuropsychologist) on six participants with aphasia who were able to communicate. Our work contributes (1) evidence that the functions implemented in the Aphasia Create tablet app suit the needs of target users, but older people are often not familiar with tactile devices, (2) reports that the Google Glass device may be problematic for persons with right-hand paresis, and (3) a characterization of the design guidelines for apps for aphasics. Both applications can be used to work with people with aphasia, and can be further developed. Aphasic centers, in which the apps were presented, expressed interest in using them to work with patients. The Aphasia Create app won the Enactus Poland National Competition in 2015.
- Open Access
Deep Learning-Based Analysis of Face Images as a Screening Tool for Genetic Syndromes
Approximately 4% of the world’s population suffers from rare diseases. A vast majority of these disorders have a genetic background. The number of genes that have been linked to human diseases is constantly growing, but there are still genetic syndromes that remain to be discovered. The diagnostic yield of genetic testing is continuously developing, and the need for testing is becoming more significant. Due to limited resources, including trained clinical geneticists, patients referred to clinical genetics units must be accurately selected. Around 30–40% of genetic disorders are associated with specific facial characteristics called dysmorphic features. As part of our research, we analyzed the performance of classifiers based on deep learning face recognition models in detecting dysmorphic features. We tested two classification problems: a multiclass problem (15 genetic disorders vs. controls) and a two-class problem (disease vs. controls). In the multiclass task, the best result reached an accuracy level of 84%. The best accuracy result in the two-class problem reached 96%. More importantly, the binary classifier detected disease features in patients with diseases that were not previously present in the training dataset. The classifier was able to generalize differences between patients and controls, and to detect abnormalities without information about the specific disorder. This indicates that a screening tool based on deep learning and facial recognition could not only detect known diseases, but also detect patients with diseases that were not previously known. In the future, this tool could help in screening patients before they are referred to the genetic unit.
Open AccessDeep Learning-Based Analysis of Face Images as a Screening Tool for Genetic Syndromes
Approximately 4% of the world’s population suffers from rare diseases. A vast majority of these disorders have a genetic background. The number of genes that have been linked to human diseases is constantly growing, but there are still genetic syndromes that remain to be discovered. The diagnostic yield of genetic testing is continuously developing, and the need for testing is becoming more significant. Due to limited resources, including trained clinical geneticists, patients referred to clinical genetics units must be accurately selected. Around 30–40% of genetic disorders are associated with specific facial characteristics called dysmorphic features. As part of our research, we analyzed the performance of classifiers based on deep learning face recognition models in detecting dysmorphic features. We tested two classification problems: a multiclass problem (15 genetic disorders vs. controls) and a two-class problem (disease vs. controls). In the multiclass task, the best result reached an accuracy level of 84%. The best accuracy result in the two-class problem reached 96%. More importantly, the binary classifier detected disease features in patients with diseases that were not previously present in the training dataset. The classifier was able to generalize differences between patients and controls, and to detect abnormalities without information about the specific disorder. This indicates that a screening tool based on deep learning and facial recognition could not only detect known diseases, but also detect patients with diseases that were not previously known. In the future, this tool could help in screening patients before they are referred to the genetic unit.
- Open Access
- 20 February 2020
- Published by MDPI AG in Diagnostics
- Vol. 10 (2) , 116
Modeling Morquio A Syndrome: An Anthropometric Study of Body Characteristics and Stature
Background: Morquio A syndrome or mucopolysaccharidosis (MPS) IVA is an autosomal recessive, life-limiting lysosomal storage disease caused by deficient activity of the enzyme galactosamine-6-sulfatase. Common early symptoms such as abnormalities of body stature can facilitate timely diagnosis. This study aimed to create a pattern of face and body stature based on anthropometric measurements taken from a cohort of Polish patients with MPS IVA. Methods: Analysis of 11 somatometric and 14 craniofacial features was performed on 20 patients with MPS IVA, aged from 3 months to 26 years. The diagnosis of MPS IVA was confirmed by enzymatic and molecular analysis. Two-tailed t-tests were used to compare mean values for body length and weight at birth between the MPS IVA patients and the general population. To show the degree and direction of deviation z-scores were calculated and then used to construct a model of an average MPS IVA patient. Results: Mean values for body height and weight at birth were greater for boys than for the general population. The observed pattern of head and body shape indicated that dwarfism occurred with age as a result of the relatively short trunk and lower limbs. Skeletal abnormalities included a bell-shaped chest with the ratio of chest depth to chest width being significantly above the norm. The head and neck were relatively elongated, in comparison to body height, and tucked between narrow shoulders. The head had dolichocephalic shape, while the nose was short with wide nostrils. Conclusions: Multiple anthropometric measurements, including age ranges, allowed for the creation of a model that showed the most characteristic features of the MPS IVA phenotype.
- 20 February 2020
- Published by MDPI AG in Diagnostics
- Vol. 10 (2) , 116
Open AccessModeling Morquio A Syndrome: An Anthropometric Study of Body Characteristics and Stature
Background: Morquio A syndrome or mucopolysaccharidosis (MPS) IVA is an autosomal recessive, life-limiting lysosomal storage disease caused by deficient activity of the enzyme galactosamine-6-sulfatase. Common early symptoms such as abnormalities of body stature can facilitate timely diagnosis. This study aimed to create a pattern of face and body stature based on anthropometric measurements taken from a cohort of Polish patients with MPS IVA. Methods: Analysis of 11 somatometric and 14 craniofacial features was performed on 20 patients with MPS IVA, aged from 3 months to 26 years. The diagnosis of MPS IVA was confirmed by enzymatic and molecular analysis. Two-tailed t-tests were used to compare mean values for body length and weight at birth between the MPS IVA patients and the general population. To show the degree and direction of deviation z-scores were calculated and then used to construct a model of an average MPS IVA patient. Results: Mean values for body height and weight at birth were greater for boys than for the general population. The observed pattern of head and body shape indicated that dwarfism occurred with age as a result of the relatively short trunk and lower limbs. Skeletal abnormalities included a bell-shaped chest with the ratio of chest depth to chest width being significantly above the norm. The head and neck were relatively elongated, in comparison to body height, and tucked between narrow shoulders. The head had dolichocephalic shape, while the nose was short with wide nostrils. Conclusions: Multiple anthropometric measurements, including age ranges, allowed for the creation of a model that showed the most characteristic features of the MPS IVA phenotype.
- Open Access
- 14 June 2019
- Published by Springer Nature in Australasian Physics & Engineering Sciences in Medicine
- Vol. 42 (3) , 701-710
Applying EMG technology in medial and lateral elbow enthesopathy treatment using Myo motion controller
Electromyography (EMG) is a diagnostic technique allowing for the detection of signals generated by changes in electrical potentials of striated muscles. The application of this technology is becoming an increasingly popular subject of scientific research. With the appearance of new devices retrieving EMG data, novel methods of its processing for various purposes are being developed. One such device is the Myo movement controller, produced by Thalmic Labs (now North). The device has been used for the analysis of muscle activation levels in patients with "tennis elbow" and "golfer’s elbow"—conditions of upper limbs which usually result from occupational injuries. The process of their rehabilitation is complex and requires a continuous monitoring of its progress. The data obtained by means of the Myo controller was used for pattern recognition of an injured hand with relation to the healthy one. The study involved examining ten subjects, including five controls. The results indicate that the muscle activation force is considerably lower in injured individuals. The arithmetic mean for the 6 analyzed motions in the injured group is 38.54% lower. The SmartEMG application ( https://www.smartemg.com ) enables the implementation of procedures performed during an examination as well as those involved in the management of the collected recordings. The study produced satisfactory results, which indicates the possibility of using the Myo controller in the treatment of elbow enthesopathy.
- 14 June 2019
- Published by Springer Nature in Australasian Physics & Engineering Sciences in Medicine
- Vol. 42 (3) , 701-710
Open AccessApplying EMG technology in medial and lateral elbow enthesopathy treatment using Myo motion controller
Electromyography (EMG) is a diagnostic technique allowing for the detection of signals generated by changes in electrical potentials of striated muscles. The application of this technology is becoming an increasingly popular subject of scientific research. With the appearance of new devices retrieving EMG data, novel methods of its processing for various purposes are being developed. One such device is the Myo movement controller, produced by Thalmic Labs (now North). The device has been used for the analysis of muscle activation levels in patients with "tennis elbow" and "golfer’s elbow"—conditions of upper limbs which usually result from occupational injuries. The process of their rehabilitation is complex and requires a continuous monitoring of its progress. The data obtained by means of the Myo controller was used for pattern recognition of an injured hand with relation to the healthy one. The study involved examining ten subjects, including five controls. The results indicate that the muscle activation force is considerably lower in injured individuals. The arithmetic mean for the 6 analyzed motions in the injured group is 38.54% lower. The SmartEMG application ( https://www.smartemg.com ) enables the implementation of procedures performed during an examination as well as those involved in the management of the collected recordings. The study produced satisfactory results, which indicates the possibility of using the Myo controller in the treatment of elbow enthesopathy.
- Open Access
- 2 April 2019
- Published by Elsevier in International Journal of Pediatric Otorhinolaryngology
- Vol. 122, 82-88
Electroglottographic and acoustic analysis of voice in children with vocal nodules
- 2 April 2019
- Published by Elsevier in International Journal of Pediatric Otorhinolaryngology
- Vol. 122, 82-88
Open AccessElectroglottographic and acoustic analysis of voice in children with vocal nodules