Electrolarynx Research Articles

Techniques chirurgicales de réhabilitation vocale après laryngectomie totale

Total laryngectomy remains indicated in the treatment of extended invasive squamous cell carcinomas of the larynx; its major consequence is the loss of voice. For more than 20 years, numerous specialists work on voice rehabilitation following total laryngectomy. Rehabilitation techniques include oesophageal speech, artificial larynx, tracheo-oesophageal shunt operations and, more recently, surgical restoration of voice with prosthesis. Several techniques have been described for the restoration of voice, such as the surgical technique by button voice prosthesis and the tracheo-oesophageal fistula. Compared to button voice prosthesis the tracheo-oesophageal fistula induces more important technical requirements and complications but less cost. We associated the extended hemi-laryngectomy of Pearson nearest speech reconstruction than sub total laryngectomy.

Clatter Reduction for Electronic Artificial Larynx

After total laryngectomy the normal voice can be replaced by an electronic artificial larynx (AL). However, the results of the surrogate voice are not overall satisfying due to a robotic clattered sound of that prosthesis. In this paper the outcome of a sound investigation for an AL is presented. This work is part of our AL research project on laryngectomee speech enhancement. As a result, an autoregressive model of the hearing process is derived that can directly be used for a novel speech-enhancement procedure. The model is derived in the framework of linear prediction theory based on the fact that speech can be modelled as the output of a linear filter excited by a periodic pulse train and random noise. Additionally, we developed an innovative training unit for novice laryngectomees measuring the relative inter-line energy of the speech.

Visualizing vowel‐production mechanism using simple educational tools

To develop intuitive and effective methods for educating Acoustics to students of different ages and from varied backgrounds, Arai [J. Phonetic Soc. Jpn. 5, 31–38, (2001)] replicated Chiba and Kajiyama’s physical models of the human vocal tract as educational tools and verified that the physical models and sound sources, such as an artificial larynx, yield a simple but powerful demonstration of vowel production in the classroom. We have also started exhibiting our models at the Science Museum ‘‘Ru‐Ku‐Ru’’ in Shizuoka City, Japan. We further extended our model to a lung model as well as several head‐shaped models with visible vocal tract to demonstrate the total vowel‐production mechanism from phonation to articulation. The lung model imitates the human respiratory system with a diaphragm. In the head‐shaped model, the midsaggital cross section is visible from the outside. To adjust the degree of nasopharyngeal coupling, the velum may be rotated. Another head‐shaped model with the manipulable tongue position was also developed. Two test results were compared before and after using these physical models, and the educational effectiveness of the models was confirmed. The homepage of the vocal‐tract models is available at http://www.splab.ee.sophia.ac.jp/Vocal—Tract—Model/index‐e.htm. [Work supported by KAKENHI (17500603).]

Analysis by synthesis of electrolarynx speech

The ability to accurately synthesize electrolarynx (EL) speech may provide a basis for better understanding the acoustic deficits that contribute to its poor quality. Such information could also lead to the development of acoustic enhancement methods that would improve EL speech quality. This effort was initiated with an analysis-by-synthesis approach that used the Klatt formant synthesizer to study vowels at the end of utterances spoken by the same subjects both before and after laryngectomy (normal versus EL speech). The temporal and spectral features of the original speech waveforms were analyzed and the results were used to guide synthesis and to identify parameters for modification. EL speech consistently displayed utterance-final fixed (mono) pitch and normal-like falling amplitude across different vowels and subjects. Subsequent experiments demonstrated that it was possible to closely match the acoustic characteristics and perceptual quality of both normal and EL speech with synthesized replicas. It was also shown that the perceived quality of the synthesized EL speech could be improved by modification of pitch parameters to more closely resemble normal speech. Some potential approaches for modifying the pitch of EL speech in real time will be discussed. [Funded by NIH Grant R01 DC006449.]

Impact of Aberrant Acoustic Properties on the Perception of Sound Quality in Electrolarynx Speech

A large percentage of patients who have undergone laryngectomy to treat advanced laryngeal cancer rely on an electrolarynx (EL) to communicate verbally. Although serviceable, EL speech is plagued by shortcomings in both sound quality and intelligibility. This study sought to better quantify the relative contributions of previously identified acoustic abnormalities to the perception of degraded quality in EL speech. Ten normal listeners evaluated the sound quality of EL speech tokens that had been acoustically enhanced by (a) increased low-frequency energy, (b) EL-noise reduction, and (c) fundamental frequency variation to mimic normal pitch intonation in relation to nonenhanced EL speech, normal speech, and normal monotonous speech (fundamental frequency variation removed). In comparing all possible combinations of token pairs, listeners were asked to identify which one of each pair sounded most like normal natural speech, and then to rate on a visual analog scale how different the chosen token was from normal speech. The results indicate that although EL speech can be most improved by removing the EL noise and providing proper pitch information, the resulting quality is still well below that of normal natural speech or even that of monotonous natural speech. This suggests that, in addition to the widely acknowledged acoustic abnormalities examined in this investigation, there are other attributes that contribute significantly to the unnatural quality of EL speech. Such additional factors need to be clearly identified and remedied before EL speech can be made to more closely approximate the sound quality of normal natural speech.

Alaryngeal Speech Enhancement Using Pattern Recognition Techniques

An alaryngeal speech enhancement system is proposed to improve the intelligibility and quality of speech signals generated by an artificial larynx transducer (ALT). Proposed system identifies the voiced segments of alaryngeal speech signal, by using pattern recognition methods, and replaces these by their equivalent voiced segments of normal speech. Evaluation results show that proposed system provides a fairly good improvement of the quality and intelligibility of ALT generated speech.

Vowel Effect on Glottal Parameters and the Magnitude of Jaw Opening

This study investigated the relationship among the magnitude of jaw opening, intrinsic fundamental frequency (F0), and glottal parameters in natural speech. Acoustic, jaw opening, and electroglottographic (EGG) signals were simultaneously recorded. The subjects were 10 healthy men with New Zealand English as their native language. Subjects were asked to repeat a standard nonemphasized sentence in which one of the target vowels (/a/, /e/, /i/, /o/, and /u/) was embedded in various contexts. The glottal parameters F0, open quotient (OQ), and speed quotient (SQ) were measured from the EGG signal. Results of a series of one-way repeated-measures analyses of variance (ANOVA) showed a significant vowel effect on the magnitude of jaw opening [F(4, 24) = 25.512, P < .001], F0 [F(4, 28) = 45.415, P < .001] and speed quotient [F(4, 28) = 5.233, P = .003], but not on the open quotient [F(4, 28) = 0.501, P = .735]. The magnitude of jaw opening was found to be inversely related with F0 (r = -0.624, n = 25, P = .0009). These findings showed that the magnitude of jaw opening was related to F0 and that jaw opening might be a control signal for simulation of long-term F0 variation to achieve a higher degree of naturalness in artificial voice.

Sultan of sound [speech technology

This paper describes the pioneering research in the field of speech technology by James L. Flanagan, 2005 IEEE Medal of Honor awardee. Flanagan's work with speech coding heralded a series of advances over the years, including a currently favored technique, linear predictive coding. After graduating from Mississippi State as an electrical engineering major, Flanagan accepted a graduate assistantship in MIT's acoustics lab, which led to his seminal research in voice coding. Flanagan then worked at Bell Telephone Laboratories where he would spend the next 33 years. He climbed steadily up the ranks at Bell Labs, eventually becoming director of the Information Principles Research Laboratory. Among the projects that Flanagan was deeply involved in were the development of automatic speech recognition systems, voice mail, artificial larynx, and packet-switched voice technology.

Objective Quality Assessment of Wideband Speech Coding

Recently, wideband speech communication using 7 kHz-wideband speech coding, as described in ITU-T Recommendations G.722, G.722.1, and G.722.2, has become increasingly necessary for use in advanced IP telephony using PCs, since, for this application, hands-free communication using separate microphones and loudspeakers is indispensable, and in this situation wideband speech is particularly helpful in enhancing the naturalness of communication. An objective quality measurement methodology for wideband-speech coding has been studied, its essential components being an objective quality measure and an input test signal. This paper describes Wideband-PESQ conforming to the draft Annex to ITU-T Recommendation P.862, Perceptual Evaluation of Speech Quality (PESQ), as the objective quality measure, by evaluating the consistency between the subjectively evaluated MOS (Mean Opinion Score) and objectively estimated MOS. This paper also describes the verification of artificial voice conforming to Recommendation P.50 Artificial Voices, as the input test signal for such measurements, by evaluating the consistency between the objectively estimated MOS using a real voice and that obtained using an artificial voice.

1A1-N-043 人工喉頭を用いたSinger Robotの研究 : 母音を連続生成する調音機構(アミューズメントロボットエンタテイナーロボット,生活を支援するロボメカ技術のメガインテグレーション)

1A1-N-043 人工喉頭を用いたSinger Robotの研究 : 母音を連続生成する調音機構(アミューズメントロボットエンタテイナーロボット,生活を支援するロボメカ技術のメガインテグレーション)

A realtime clarification filter of dysphonic speech and its evaluation by listening experiments

This paper presents a digital filtering algorithm which clarifies dysphonic speech with the speaker’s individuality preserved. The study deals with the clarification of oesophageal speech and the speech of patients with cerebral palsy, and the filtering ability is being evaluated by listening experiments. Over 20,000 patients are currently suffered from laryngeal cancer in Japan, and the only treatment for the terminal symptoms requires the removal of the larynx including vocal cords. The authors are developing a clarification filtering algorithm of oesophageal speech, and the primal algorithm of software clarification and its effectiveness was reported in the previous ICDVRAT. Several algorithms for the clarification have been newly developed and implemented, and are being evaluated by questionnaires. The algorithms were extended and applied for the clarification of the speech by the patients of cerebral palsy. A voice is the most important and effective medium employed not only in the daily communication but also in logical discussions. Only humans are able to use words as means of verbal communication, although almost all animals have voices. Vocal sounds are generated by the relevant operations of the vocal organs such as a lung, trachea, vocal cords, vocal tract, tongue and muscles. The airflow from the lung causes a vocal cord vibration to generate a source sound, then the glottal wave is led to the vocal tract, which works as a sound filter as to form the spectrum envelope of a particular voice. If any part of the vocal organs is injured or disabled, we may be involved in the impediment in the vocalization and, in the worst case, we may lose our voices. Over 20,000 patients are currently suffered from laryngeal cancer in Japan, and the only treatment for the terminal symptoms is to remove the larynx. The removal of vocal cords means the loss of the voice, and causes various difficulties in the communication with other people, since the employment of a voice is essentially important for humans to make verbal communications. There are mainly two ways to recover voice. One is to use an artificial larynx, which is a hand-held device with a pulse generator that produces a vocal cord-like vibration. An electrolarynx has a vibrating plastic diaphragm, which is placed against the neck during the speech. The vibration of the diaphragm generates a source sound in the throat, and the speaker then articulates with the tongue, palate, throat and lips as he does for the usual vocalization. The device has an advantage to be used by just being held to the neck and to be easily mastered, but the sound quality is rather electronic and artificial. Furthermore one hand is occupied to hold the device during the speech, which disturbs the gestural communication. The other way is to train oesophageal speech, which is a method of speech production using an oesophagus (Sato, 1993; Max et al, 1996). In the speech, air is inhaled and caught in the upper oesophagus instead of being swallowed, and then the released air generates the oesophagus vibration to produce a “belch-like” sound that can be shaped into speech. A patient has difficulties to master the oesophageal speech (several years are ordinary required for the practice), however the voice is able to keep the speaker’s individuality since the speech is generated by his own vocal organs, although several distinctive characteristics exist. Moreover the speaker is able to employ his body parts such as hands and facial expressions freely and actively for the communication to assist the speech. Cerebral palsy (CP) is a condition caused by an injury to the parts of the brain which controls the ability to use muscles and bodies. The injury may happen before birth, sometimes during delivery, or soon after

At the Listening Post, or, do machines perform?

Listening Post, a collaboration between statistician Mark Hansen and media artist Ben Rubin, samples conversations from public Internet chat sites. The sampled texts are displayed on small LCD screens while an artificial voice reads them aloud. The installation also generates music algorithmically from the data it samples. There are many ways one could describe or categorize Listening Post: as a highly specialized computer, as an electronic sculpture, or as an installation, for example. In some of my recent work I consider the implications of calling such pieces performances. To create a context for a discussion of Listening Post, I briefly summarize some of my thinking on performances by machines with particular attention to the questions raised by identifying machines as performers and the issue of liveness in performance.

Intrauterine Cerebral Processing of External Vibroacoustic Stimuli

The human and the sheep fetus, the animal model of human pregnancy, respond to external stimuli such as vibrations or noises with behavior state changes from 0.8 gestation onwards. It is not clear yet 1) if the behavior state changes that represent a „subcortical“ arousal are accompanied by a „cortical“ arousal as a condition of conscious perception of the stimuli; 2) to what extent the perception of the stimuli depends on the behavior state and the quality of the stimulus; and 3) if the proprioceptive system is important for perception of the vibratory component of the signal. Via Cesarean section, sheep fetuses were chronically instrumented at 0.8 gestation with ECoG (electrocorticogram) electrodes as well as EOG, EMG and EKG electrodes to define fetal behavior states. After a recovery period of five days, stimuli were presented from the abdominal wall of the ewe during periods of NREM and REM sleep. We tested pure vibratory (30min) and acoustic (5h) stimuli as well as dynamic amplitude and frequency modulated and two static vibroacoustic stimuli (VAS) produced by an electronic artificial larynx (EAL) and a laboratory shaker. ECoG changes were analyzed using linear (power spectral analysis) and non-linear methods. For non-linear analysis we used an algorithm based on the Wolf algorithm (calculation of the leading Lyapunov exponent) which calculates a point prediction error (PE) regarding the course of the time series in the phase space. The vibratory and acoustic stimuli did not induce a cortical arousal. The VAS produced an arousal with similar effectiveness during NREM and REM sleep. The dynamic stimulus induced 100%, the EAL 70% and the laboratory shaker 40–50% responses. The PE showed a high predictability in NREM and a low predictability of the ECoG in REM sleep. Although the power spectrum differed between VAS in NREM (decrease of delta activity) and REM sleep (beta activation), the PE showed that the cortical activation pattern did not depend on the sleep state and the stimulus. Processes of cortical activation and deactivation during spontaneous sleep state changes and VAS were similar in their time courses. Cochlear ablation prevented from arousal, suggesting that stimuli were detected through the auditory rather than the proprioceptive system. The results show that external VAS induce a cortical arousal during the last trimester depending on the stimulus. The time course of cortical activation within 32s suggests an arousal via the ascending activating reticular system rather than via specific fiber systems.

Advances in Alaryngeal Communication and the Art of Tracheoesophageal Voice Restoration

Advances in Alaryngeal Communication and the Art of Tracheoesophageal Voice Restoration

A constructivist approach to infants' vowel acquisition through mother–infant interaction

Inspired by the observation that infants acquire phonemes common to adults without having the capability to articulate, nor having a priori knowledge about the relationship between the sensorimotor system and phonemes, a constructivist approach to building a robot that reproduces a similar developmental process is conducted. Two general issues are addressed: what are the interactive mechanisms involved and what should be the behaviour of the caregiver/teacher? Based on findings in developmental psychology, it is conjectured that: (a) the caregiver's vocalization in response to infants' cooing reinforces the infant's articulation along the caregiver's phonemic categories; and (b) the caregiver's repetition with adult phonemes helps to specify the correspondence between cooing and the caregiver's phonemes as well as determining the acoustic properties of the phonemes. The robot consists of an artificial articulatory system with a five-degrees of freedom mechanical system deforming a silicon vocal tract connected to an artificial larynx, an extractor of formants, and a learning mechanism with self-organizing auditory and articulatory layers. Starting off with random vocalizations, the system uses the caregiver's repetitive utterances to bootstrap its learning. In order to resolve the arbitrariness in determining proper articulations, the torque to deform the tract and its resultant deformation are minimized. The experimental results, discussion and future issues are given.

A simulation-based methodology and tool for automating the modeling and analysis of voice-over-IP perceptual quality

A simulation-based methodology is developed for analyzing the perceptual quality of voice-over-IP calls as a function of network quality-of-service (QoS) parameters and choices in configuration and implementation. The proposed method combines the use of existing objective voice-quality measurement algorithms, such as the ITU-T P.861 PSQM and P.862 PESQ, and pre-recorded natural or artificial voice reference signals, such as the ITU-T P.50, with the discrete-event simulation of a network QoS model. A significant advantage of the method is that it does not involve the use of human subjects in evaluating subjective voice quality. This enables one to entirely automate the process of quantifying call quality as a function of network QoS and implementation choices such as packet size and codec type. Such automation enables one to realize significant time and cost savings in obtaining experimental results. A tool implementation is described that includes basic network packet loss, delay jitter, and call multiplexing models. Example numerical results are presented. The generalization of the method to the automated perceptual quality evaluation of audio, video, and multimedia signals is also described.

Professor Nicolas Taptas (1871-1955): a pioneer of post-laryngectomy voice rehabilitation.

To report on the career of Professor Nicolas Taptas of Constantinople (1871-1955) and his contribution to the development of an artificial larynx. Historical review. The unpublished documents of Taptas's family archives and one of his papers, describing his own original technique for voice rehabilitation after total laryngectomies, were studied. In his texts, Taptas described the application of external laryngeal prostheses in one of his patients who underwent total laryngectomy. Taptas meticulously recorded his own first successful attempt in a woman with cancer of the larynx. He was one of the first scientists who, at the end of the 19th century, devised an external laryngeal prosthesis, which had the advantage of rehabilitating a sufficiently strong whispering voice; by using a prototype valve, he avoided problems of aspiration of liquids and food to the bronchi. Nicolas Taptas's contribution to the development of the artificial larynx was vital because he devised and applied one of the first external laryngeal artificial prostheses with satisfactory functional results.

Improvement in PZT Ceramic Vibrator Low Frequency Of Generated Sound for Using PZT Ceramic Vibrator As a Sound Source for an Artificial Larynx

Speech production substitutes to reconstruct speech functions have been developed and used practically to vocalize speech of those who have lost voice functions, but have problems. We focused on PZT ceramic vibrators as a sound source for an artificial larynx. We first produced the artificial larynx, which uses PZT ceramic vibrator; and then evaluated its performance. The voice of the power spectrum of the artificial larynx user is similar to that of the speaking person's voice. The vocalized sound of the artificial larynx user shows good characteristics at format frequency, which is important for vowel discrimination. The resonant frequency of PZT ceramic vibrator is high, however and it is difficult to generate sound near 125Hz for PZT ceramic vibrators. We studied how to lower the frequency of generated sound by using added mass.

Differential Modes of Alaryngeal Communication and Long-Term Voice Outcomes following Pharyngolaryngectomy and Laryngectomy

Patterns of vocal rehabilitation for 37 pharyngolaryngectomy patients and 55 total laryngectomy patients over a 5-year period were compared. An electrolarynx (EL) was introduced as the initial communication mode immediately after surgery for 98% of patients, with 30% of pharyngolaryngectomy and 74% of laryngectomy patients subsequently developing tracheoesophageal speech (TES) as their primary mode of communication. Follow-up with 14 of 37 pharyngolaryngectomy patients and 36 of 55 laryngectomy patients was conducted 1–6 years following surgery and revealed that 90% of the pharyngolaryngectomy patients maintained the use of TES in the long term compared to 69% of the laryngectomy group. Long-term outcomes relating to communication disability and handicap did not differ significantly between the two surgical groups, however the laryngectomy patients had significantly higher levels of wellbeing. Across the whole group of patients, statistical comparison revealed that patients using TES had significantly lower levels of disability, handicap and distress than EL users. Considering that lower levels of disability, handicap and distress are associated with TES, and the data supports that suitably selected patients can maintain functional TES in the long term, increased application of this form of communication rehabilitation should be encouraged where viable for the pharyngolaryngectomy population.

Voice restoration with low pressure blom singer voice prosthesis after total laryngectomy.

The main problem after total laryngectomy is permanent loss of voice. Current methods of vocal rehabilitation after total laryngectomy include development of esophageal speech, use of artificial larynx, tracheoesophageal shunt operations and more recently surgical restoration of the voice with prosthesis. Primary voice restoration using Blom- Singer voice prosthesis after total laryngectomy and pharyngeal myotomy was performed in 187 patients between October 1992 and July 2000. There were 184 male and 3 female patients of average age 63.7 years (range 42-76). Mean follow up period was 62 months. Satisfactory speech was achieved in 156 patients (83.5%). During the follow-up period, we experienced complaints of insufficient voice in 31 (16.5%) patients, due to partial spasm in 17 and total spasm in the pharyngoesophageal segment in 14. Furthermore, 24 (12.8%) patients preferred esophageal speech or electro larynx because of low socioeconomic level. The overall success rate was 70.7%. In this study the results of the surgical technique and prosthesis insertion, as well as the associated complications and socioeconomic levels of the patients, are discussed.