Electrolarynx Research Articles

Self‐Powered Wearable Pressure Sensors for Detection and Separation of Signals for Various Human Movements

A detailed study on the dynamic response of mountable pressure sensors is presented, with a focus on foot pressure sensors integrated with carbon nanotube (CNT)‐coated cotton fibers. The research explores the sensor's sensitivity to pressure changes, repeatability, hysteresis, and durability through rigorous modeling and experimental validation. Computational simulations using Python and experimental data demonstrate the sensor's nonlinear conductance response to applied force, attributed to the varying contact area and number of contact points among the fibers. Long‐term outdoor exposure tests confirm the material's resilience, maintaining electrical conductivity and structural integrity. The study also demonstrates the sensor's ability to monitor human activities like walking, running, stair climbing, and jumping, as well as detecting muscle movements during swallowing, coughing, and speech, highlighting potential applications in health monitoring and artificial voice synthesis. The MRMR algorithm is used for feature selection to distinguish activities, demonstrating the sensor's potential for activity recognition. Additionally, a piezoelectric sensor on the pressure sensors estimates harvested electric power to power wearable devices. This work contributes to the advancement of self‐powered wearable pressure sensors to monitor real‐time human activity, with implications for healthcare, sports performance, and assistive technologies.

CNN-Based Object Recognition and Tracking System to Assist Visually Impaired People

Visually impaired peoples (VIPs) are a significant population segment worldwide. A system has been developed to aid mobility and ensure their safety. The system uses artificial voice for real-time navigation, allowing VIPs to perceive their surroundings. A web application is also created for their safety, allowing them to share their location with family. The project focuses on object recognition, converting detected object names (such as person1, person2, Dog, cat) into audio for easier understanding. Keywords— Visually impaired peoples (VIPs), artificial voice, object recognition, object name, audio

Effect of Curvature Shape of Transparent COVID-19 Protective Face Shields on the Speech Signal

Recent papers and studies over the course of last three years have shown that COVID-19 has a negative impact on the speech communication quality between people. This paper presents an influence analysis of the curvature shape of protective transparent shields on the speech signal. Five shields made of the same material and dimensions but with different curvatures were analyzed, from a completely flat to a very curved shield which has the same shape of curvature at its top and bottom and covers the entire face. The influence of the shield is analyzed with two types of experiments – one using dummy head with integrated artificial voice device, and the other using real speakers (female and male actors). It has been shown that usage of protective shields results in a relative increase in the speech signal level, in the frequency range of around 1000 Hz, compared to the situation when protective shields are not used. The relative increase in speech signal levels for large-curvature shields can be up to 8 dB. The possible causes of this phenomenon have been analyzed and examined.

Design of smart home environment based on wireless sensor system and artificial speech recognition

In the context of the rapid development of intelligent technology, smart home has become the focus of attention. However, the traditional smart home system has some problems such as complicated operation and inconvenient information interaction. In order to solve these problems and improve the user experience of smart home environment, this study establishes a fast and efficient smart home environment design scheme by introducing wireless sensor system and artificial speech recognition technology. By arranging multiple wireless sensor nodes, the monitoring and data acquisition of environmental parameters are realized, and then the voice control and command interaction of smart home devices are realized by using artificial voice recognition technology. Through the effective combination of wireless sensor system and artificial voice recognition technology, users can realize the control of smart home environment through voice commands, and improve the convenience and intelligence of the system. The experimental results show that the design scheme can accurately and stably realize the recognition of user voice commands and the control of smart home devices, and users' evaluation of the degree of intelligence and user experience has also been significantly improved.

Voice Synthesis Improvement by Machine Learning of Natural Prosody.

Since the advent of modern computing, researchers have striven to make the human-computer interface (HCI) as seamless as possible. Progress has been made on various fronts, e.g., the desktop metaphor (interface design) and natural language processing (input). One area receiving attention recently is voice activation and its corollary, computer-generated speech. Despite decades of research and development, most computer-generated voices remain easily identifiable as non-human. Prosody in speech has two primary components-intonation and rhythm-both often lacking in computer-generated voices. This research aims to enhance computer-generated text-to-speech algorithms by incorporating melodic and prosodic elements of human speech. This study explores a novel approach to add prosody by using machine learning, specifically an LSTM neural network, to add paralinguistic elements to a recorded or generated voice. The aim is to increase the realism of computer-generated text-to-speech algorithms, to enhance electronic reading applications, and improved artificial voices for those in need of artificial assistance to speak. A computer that is able to also convey meaning with a spoken audible announcement will also improve human-to-computer interactions. Applications for the use of such an algorithm may include improving high-definition audio codecs for telephony, renewing old recordings, and lowering barriers to the utilization of computing. This research deployed a prototype modular platform for digital speech improvement by analyzing and generalizing algorithms into a modular system through laboratory experiments to optimize combinations and performance in edge cases. The results were encouraging, with the LSTM-based encoder able to produce realistic speech. Further work will involve optimizing the algorithm and comparing its performance against other approaches.

Factors associated with successful electrolarynx use after total laryngectomy, a multi-institutional study.

To identify characteristics associated with successful electrolarynx (EL) use after total laryngectomy (TL). Records of 196 adults who underwent TL from 03/15/2012 to 03/15/2022 at the University of Washington and Puget Sound Veterans Affairs were reviewed. Characteristics included age, Charlson Comorbidity Index, social support, pre-operative radiation (RT) and chemoradiation (CRT), and 6-month post-TL swallow status. EL success was evaluated using pre-defined criteria of intelligibility, reliability, and independence with use. Poisson regressions and robust standard error estimates were used to estimate unadjusted risk ratios for each characteristic. Statistically significant characteristics were included in multivariate analysis (MVA) to estimate adjusted risk ratios. Median age was 64, median Charlson Comorbidity Index was 5, 170 (87%) were male, 159 (81%) had high social support, and 159 (81%) attained post-TL full-oral diet. Pre-operatively, 110 (56%) had RT, including 55 (28%) with CRT. Ninety-three (47%) met our criteria for EL success. Characteristics significantly associated with EL success included social support (p = .037) and post-TL full-oral diet (p = .037); both approached significance on MVA. EL success varied by pre-operative treatment on univariate (p = .005) and MVA (p = .014). Compared to no prior RT or CRT, the probability of EL success was 29% higher with prior RT and 29% lower with prior CRT in MVA, although these associations did not reach significance. In this retrospective review, EL success correlated with high social support, post-TL full-oral diet, and pre-operative treatment history. These results warrant validation in a larger prospective study to help guide the choice of voice rehabilitation modalities or intensified speech therapy. 4.

Voice Cloning Using Artificial Intelligence and Machine Learning: A Review

This paper represents a thorough method for integrating emotions, texttospeech conversion, and state of the art voice cloning. The paper focuses on novel background noise adaptation, emotional voice synthesis, and multi-speaker voice cloning for better speech synthesis. The synthesis of emotive voices, multi-speaker voice cloning, and creative methods for modifying background noise to improve speech synthesis quality are among the topics covered in this study. Additionally, the study explores the domain of emotional artificial intelligence by adding a variety of emotions to artificial voices, improving user engagement through sympathetic reactions. The study also looks at how background noise can be altered to change it from a disturbing to a silent, non-disruptive state. The texttospeech systems usability in noisy conditions is greatly enhanced by this improvement. By integrating these components, the project makes a substantial contribution to text to speech, emotional AI, and voice cloning, creating new avenues for human-computer connection.

Matching digital companions with customers: The role of perceived similarity

AbstractDigital companions are an advanced form of digital agents that do not only provide advice and support but accompany people on their day‐to‐day customer journeys. This article sheds light on the psychological processes underlying customers’ responses to these digital companions (i.e., virtual friends or co‐consumers). We propose that framing them as matched with customers on goal‐relevant attributes (i.e., attributes related to customers’ consumption goals) fosters positive customer outcomes (i.e., consumption enjoyment and positive word‐of‐mouth), mediated by perceived similarity in these attributes. Importantly, in this matching context, humanlikeness serves as a boundary condition for perceived similarity to occur. Furthermore, the effect of perceived similarity on customer outcomes is driven by perceived connectedness. In Study 1, in the context of experiential learning, we identified shared interest and personality as goal‐relevant attributes underlying perceived similarity. With the manipulation of the match frame and humanlike versus artificial voice of the digital companion, Study 2 supports our propositions and highlights shared interest, but not personality, as the core driver. We provide recommendations on how to design and market digital companions to foster connection and favorable customer outcomes.

Patterns of alaryngeal voice adoption and predictive factors of vocal rehabilitation failure following total laryngectomy.

We aimed to determine patterns of alaryngeal voice acquisition and predictive factors of vocal rehabilitation (VR) failure following total laryngectomy (TL) at a large Canadian tertiary care center. All consecutive patients having undergone a TL between January 1st, 2011 and December 31st, 2019, at the Centre Hospitalier de l'Université de Montréal were included. One hundred and ninety-seven laryngectomized patients were identified. Successful VR was achieved in 86 (59.0%) patients, while 59 (41.0%) failed to use a method of alaryngeal voice as their principal means of communication at 1 year postoperatively. The use of tracheoesophageal puncture (TEP) was associated with higher VR success rates (70.6%) when compared with the artificial larynx (48.6%), and esophageal voice (18.8%). The only independent predictor of VR failure on multivariate analysis at all time points was a low socioeconomic status. Failure to adopt an alaryngeal voice following TL is highly prevalent, despite comprehensive and free speech language pathologist services being offered at our center. A low resort to TEP at our institution and a poor acceptability and accessibility of alternative VR methods may contribute to this trend. The challenges of VR may be further exacerbated by the barriers linked to a lower socioeconomic status, which in turn may contribute to reduced candidacy for TEP.

Effects of vocal fold oscillation characteristics on the aerosol droplet production

Although a large number of aerosol droplets are known to be produced from the human larynx during vocalizations, it remains unclear how vocal fold oscillations are involved in the droplet production. In this study, we synthesized vocalizations using two types of silicone vocal fold models and investigated the effects of these model configurations on the droplet production mechanisms by measuring the particle concentrations when artificial mucus was injected into the artificial larynx. One model, which oscillates with complete glottal closure, produced louder sounds with droplets, whose diameter ranged primarily from 0.3 to 2μm. In contrast, the other model, which oscillates without vocal fold collision, produced smaller sounds with droplets having larger diameters. Additional observations using the laser sheets and high-speed imaging demonstrated that the liquid films were formed just above the vocal folds when a constant jet flow was generated in the model which does not exhibit complete glottal closure. These results describe how different oscillation characteristics of the vocal folds may cause differences in droplet concentration.

Warning: Humans cannot reliably detect speech deepfakes.

Speech deepfakes are artificial voices generated by machine learning models. Previous literature has highlighted deepfakes as one of the biggest security threats arising from progress in artificial intelligence due to their potential for misuse. However, studies investigating human detection capabilities are limited. We presented genuine and deepfake audio to n = 529 individuals and asked them to identify the deepfakes. We ran our experiments in English and Mandarin to understand if language affects detection performance and decision-making rationale. We found that detection capability is unreliable. Listeners only correctly spotted the deepfakes 73% of the time, and there was no difference in detectability between the two languages. Increasing listener awareness by providing examples of speech deepfakes only improves results slightly. As speech synthesis algorithms improve and become more realistic, we can expect the detection task to become harder. The difficulty of detecting speech deepfakes confirms their potential for misuse and signals that defenses against this threat are needed.

An acoustic study of Cantonese alaryngeal speech in different speaking conditions.

Esophageal (ES) speech, tracheoesophageal (TE) speech, and the electrolarynx (EL) are common methods of communication following the removal of the larynx. Our recent study demonstrated that intelligibility may increase for Cantonese alaryngeal speakers using clear speech (CS) compared to their everyday "habitual speech" (HS), but the reasoning is still unclear [Hui, Cox, Huang, Chen, and Ng (2022). Folia Phoniatr. Logop. 74, 103-111]. The purpose of this study was to assess the acoustic characteristics of vowels and tones produced by Cantonese alaryngeal speakers using HS and CS. Thirty-one alaryngeal speakers (9 EL, 10 ES, and 12 TE speakers) read The North Wind and the Sun passage in HS and CS. Vowel formants, vowel space area (VSA), speaking rate, pitch, and intensity were examined, and their relationship to intelligibility were evaluated. Statistical models suggest that larger VSAs significantly improved intelligibility, but slower speaking rate did not. Vowel and tonal contrasts did not differ between HS and CS for all three groups, but the amount of information encoded in fundamental frequency and intensity differences between high and low tones positively correlated with intelligibility for TE and ES groups, respectively. Continued research is needed to understand the effects of different speaking conditions toward improving acoustic and perceptual characteristics of Cantonese alaryngeal speech.

Differences in arousal and valence on the Korean phoneme of artificial voice between Korean and Chinese women.

Although active research is in progress in the fields of psychology and linguistics on the emotional characteristics of the symbol and meaning of sound itself, since the systematic emotional model is not applied, each researcher uses a subjective concept and acts as an obstacle to the expansion of research. There is a limitation in that it cannot be confirmed whether the sound symbol has universality regardless of cultural differences between different languages. In this study, the difference between the arousal and valence of emotions felt toward Hangul phonemes was explored according to consonant and vowel through comparison between Korean and Chinese women. 38 Korean women and 32 Chinese women were recruited, and an online experiment was conducted in which arousal and valence were reported for 42 Hangeul phoneme sound stimuli. As a result of comparing the arousal and valence of each group, Koreans showed significantly higher arousal scores than Chinese, and these results showed different differences according to consonant and vowel. In valence, there was a difference between nationalities only according to consonant indicating that Koreans showed lower positivity toward aspirated sounds than Chinese. Through these results, it was confirmed that the emotional meaning of the sound symbol between different languages is different, which can be affected by consonant and vowels. This study identified differences in emotional perception between cultures by using two dimensions of emotions, arousal, and valence, which are systematized for sound symbols, and suggests implications for the relationship between sound symbol and emotions and cultural differences in the future.

Subjective Perception and Psychoacoustic Aspects of the Laryngectomee Voice: The Impact on Quality of Life.

A retrospective study is presented to correlate the inter-judge consistency for the different psycho-perceptual parameters of the recently proposed Impression Noise Fluency Voicing (INFVo) perceptual rating scale for substitution voices, and the vocal function as perceived by the patient. The scale Voice-Related Quality of Life (V-RQoL) and the Self Evaluation of Communication Experiences After Laryngectomy scale (SECEL)-a self-evaluation questionnaire of communicative experience after laryngectomy surgery-were administered to 89 total laryngectomees, subdivided in four groups depending on their type of alaryngeal voice (i.e., tracheoesophageal and esophageal speakers, electro larynx users, voiceless patients), in order to evaluate the impact of the impairment of the phonatory function on the quality of life. No significant differences exist among the various groups on their perception of QoL using subjective questionnaires, whereas the INFVo scale has proven to be a useful tool for the description and analysis of the psychoacoustic characteristics of the vocal signal and a reliable instrument to correctly classify the patients. It is also notable that the judgement of the patients on their own voice and those of the referees are highly significant. Although speech rehabilitation for the acquisition of a substitution voice offers a new way of communication for the laryngectomized patients, nonetheless, their QoL is not significantly related to the type of substitution voice. Therefore, improving the patient's adaptation to the new phonatory condition is mandatory.

Deep Learning Methods for Arabic Autoencoder Speech Recognition System for Electro-Larynx Device

Recent advances in speech recognition have achieved remarkable performance comparable with human transcribers’ abilities. But this significant performance is not the same for all the spoken languages. The Arabic language is one of them. Arabic speech recognition is bounded to the lack of suitable datasets. Artificial intelligence algorithms have shown promising capabilities for Arabic speech recognition. Arabic is the official language of 22 countries, and it has been estimated that 400 million people speak the Arabic language worldwide. Speech disabilities have been one of the expanding problems in the last decades, even in kids. Some devices can be used to generate speech for those people. One of these devices is the Servox Digital Electro-Larynx (EL). In this research, we developed an autoencoder with a combination of long short-term memory (LSTM) and gated recurrent units (GRU) models to recognize recorded signals from Servox Digital EL Electro-Larynx. The proposed framework consisted of three steps: denoising, feature extraction, and Arabic speech recognition. The experimental results show 95.31% accuracy for Arabic speech recognition with the proposed model. In this research, we evaluated different combinations of LSTM and GRU for constructing the best autoencoder. A rigorous evaluation process indicates better performance with the use of GRU in both encoder and decoder structures. The proposed model achieved a 4.69% word error rate (WER). Experimental results confirm that the proposed model can be used for developing a real-time app to recognize common Arabic spoken words.

1885: First total laryngectomy and artificial larynx in France

This year, 2023, is the 150th anniversary of the first total laryngectomy for cancer, by Theodor Billroth. The authors reconstruct the conditions under which, on March 12, 1885, this operation was then performed for the first time in France, by Leon Labbé, and present the man himself, and also M. Cadier, the inventive genius who designed the first artificial larynx used in this country.

Toward a robust swallowing detection for an implantable active artificial larynx: a survey.

Total laryngectomy consists in the removal of the larynx and is intended as a curative treatment for laryngeal cancer, but it leaves the patient with no possibility to breathe, talk, and swallow normally anymore. A tracheostomy is created to restore breathing through the throat, but the aero-digestive tracts are permanently separated and the air no longer passes through the nasal tracts, which allowed filtration, warming, humidification, olfaction, and acceleration of the air for better tissue oxygenation. As for phonation restoration, various techniques allow the patient to talk again. The main one consists of a tracheo-esophageal valve prosthesis that makes the air passes from the esophagus to the pharynx, and makes the air vibrate to allow speech through articulation. Finally, swallowing is possible through the original tract as it is now isolated from the trachea. Yet, many methods exist to detect and assess a swallowing, but none is intended as a definitive restoration technique of the natural airway, which would permanently close the tracheostomy and avoid its adverse effects. In addition, these methods are non-invasive and lack detection accuracy. The feasibility of an effective early detection of swallowing would allow to further develop an implantable active artificial larynx and therefore restore the aero-digestive tracts. A previous attempt has been made on an artificial larynx implanted in 2012, but no active detection was included and the system was completely mechanic. This led to residues in the airway because of the imperfect sealing of the mechanism. An active swallowing detection coupled with indwelling measurements would thus likely add a significant reliability on such a system as it would allow to actively close an artificial larynx. So, after a brief explanation of the swallowing mechanism, this survey intends to first provide a detailed consideration of the anatomical region involved in swallowing, with a detection perspective. Second, the swallowing mechanism following total laryngectomy surgery is detailed. Third, the current non-invasive swallowing detection technique and their limitations are discussed. Finally, the previous points are explored with regard to the inherent requirements for the feasibility of an effective swallowing detection for an artificial larynx. Graphical Abstract.

AI Narratives: A History of Imaginative Thinking about Intelligent Machines ed. by Stephen Cave, Kanta Dihal, and Sarah Dillon

Reviewed by: AI Narratives: A History of Imaginative Thinking about Intelligent Machines ed. by Stephen Cave, Kanta Dihal, and Sarah Dillon Ronald R. Kline (bio) AI Narratives: A History of Imaginative Thinking about Intelligent Machines Edited by Stephen Cave, Kanta Dihal, and Sarah Dillon. Oxford: Oxford University Press, 2020. Pp. 424. The strength of this edited volume, which grew out of a series of interdisciplinary workshops organized by the Leverhulme Centre for the Future of Intelligence at the University of Cambridge and the Royal Society of London, lies in its comprehensiveness. The book's twenty contributors, representing a wide range of disciplines, mostly in the humanities, consider an impressive spectrum of technologies and time periods. Part 1 ("Antiquity to Modernity") covers clockwork, self-regulated automata, and early robots in Europe from Ancient Greece to the industrial revolutions of [End Page 223] the nineteenth and early twentieth centuries. Part 2 ("Modern and Contemporary") focuses on the electronic computer systems, including neural-net machine learning, that came to be known as "artificial intelligence" in the United States and Europe in the twentieth and twenty-first centuries. The variety of narratives is equally impressive. Part 1 deals with a wide range of texts, from Homeric epics to books on natural magic and experimental science in the Renaissance and early modern period, and from Victorian fiction to Karel Čapek's classic play R.U.R. (Rossum's Universal Robots), which premiered in 1921. Part 2, organized by themes, focuses on the rise of modern science fiction in books, films, and television, often contrasting that with futuristic predictions, both utopian and critical, made by AI practitioners. AI Narratives thus resembles a scholarly handbook that summarizes recent work in a new field of study. As with most handbooks, its strength of being comprehensive contributes to its weakness of not presenting a unifying argument or perspective about these narratives. The history and sociology of science and technology is represented in the volume by references to such well-known scholars as Jessica Riskin, Simon Schaffer, David F. Noble, N. Katherine Hayles, Judy Wajcman, and Sheila Jasanoff. But the omissions are notable too, such as Adelheid Voskuhl on automata and fiction in eighteenth-century Europe and Donna Haraway on cyborgs. Although the editors state that the "work of this book contributes to establishing the importance of narratives as constituent parts of any sociotechnical imaginary" (p. 7), a concept introduced by Jasanoff, none of the authors engage with this concept, nor with historian of technology David Nye's influential work on technological narratives. In part 2, I would have liked to have heard more about the relationship between the writers of science fiction and AI researchers. Beth Singler mentions this mutual influence in her nice chapter on AI, anthropomorphism, and the parent-child narrative. But the fact that Marvin Minsky, a founder of computer-based AI, was a consultant to the classic film 2001: A Space Odyssey (1968), or that John W. Campbell Jr., the editor of Astounding Science Fiction magazine, corresponded with his former MIT professor Norbert Wiener, a founder of cybernetics who is cited several times in AI Narratives, suggests that much more could be done to explore the permeable boundary between the creators of science fiction and the inventors of AI systems in the twentieth century. All of the chapters are well researched, well argued, and informative, whether they recount a recent book by the author or present new research. An idea of the richness of the topics and themes can be seen from the titles of some outstanding chapters. These include Minsoo Kang and Ben Halliburton's "The Android of Albertus Magnus: A Legend of Artificial Being," Julie Park's "Making the Automaton Speak: Hearing Artificial Voices in the Eighteenth Century," Kanta Dihal's "Enslaved Minds: Artificial Intelligence, Slavery, and Revolt," and Anna McFarlane's "AI and Cyberpunk Networks." [End Page 224] These virtues make AI Narratives a path-breaking book that surveys the important place of narrative in the long history of the interaction between humans and intelligent machines in Europe and the United States. Ronald R. Kline Ronald R. Kline is Bovay Professor of History and Ethics of Engineering, Emeritus, at Cornell University. He...

A New Substitute Voice Using a Smartphone Application

Although widely used by patients who have undergone laryngectomy, vocalization using an electronic artificial larynx, tracheoesophageal shunt vocalization, and esophageal speech cannot reproduce a patient’s original voice, and this negatively affects their quality of life. In 2020, our department introduced a smartphone app called “coestation®,” an AI-based software that can reproduce voices with great similarity to the patient’s original voice by synthesizing 200 voice-recorded sentences.Our user survey revealed a high level of satisfaction with the similarity between the synthesized voices and the patients’ original voices. However, patients also showed dissatisfaction with the time lag and complexity of the app. In fact, only four of the six patients who agreed to participate in the survey were able to use the app properly. Consequently, the average voice-related quality of life among the app users was 61.9%, which is slightly lower than that among patients who used traditional substitute voices. Despite the remarkable quality of the voices reproduced by this innovative app, it should be constantly improved based on the patient feedback in order to meet their expectations.

Carl Gussenbauer and the first artificial larynx after total laryngectomy

The aim of this historical vignette is to delve into the description and use of the first artificial larynx, developed by Carl Gussenbauer, and used after Theodor Billroth's first total laryngectomy, performed on December 31, 1873.