EEG-Based Auditory Attention Detection and Its Possible Future Applications for Passive BCI

The ability to discriminate and attend one specific sound source in a complex auditory environment is a fundamental skill for efficient communication. Indeed, it allows us to follow a family conversation or discuss with a friend in a bar. This ability is challenged in hearing-impaired individuals and more precisely in those with a cochlear implant (CI). Indeed, due to the limited spectral resolution of the implant, auditory perception remains quite poor in a noisy environment or in presence of simultaneous auditory sources. Recent methodological advances allow now to detect, on the basis of neural signals, which auditory stream within a set of multiple concurrent streams an individual is attending to. This approach, called EEG-based auditory attention detection (AAD), is based on fundamental research findings demonstrating that, in a multi speech scenario, cortical tracking of the envelope of the attended speech is enhanced compared to the unattended speech. Following these findings, other studies showed that it is possible to use EEG/MEG (Electroencephalography/Magnetoencephalography) to explore auditory attention during speech listening in a Cocktail-party-like scenario. Overall, these findings make it possible to conceive next-generation hearing aids combining customary technology and AAD. Importantly, AAD has also a great potential in the context of passive BCI, in the educational context as well as in the context of interactive music performances. In this mini review, we firstly present the different approaches of AAD and the main limitations of the global concept. We then expose its potential applications in the world of non-clinical passive BCI.

Acoustic characteristics of voice in voice professionals with hypotonic dysphonia

The article presents the results of a study of the acoustic characteristics of the voice in voice professionals and non-voice professionals. We examined 80 people aged 23 to 45 years with functional dysphonia of the hypotonic type, who applied to the phoniatric office of the St. Petersburg Research Institute of ENT. Of these, 23 were professional vocalists, 12 were representatives of speech professions, and 45 were non-professional voices. First, the phoniatrist examined the larynx and assessed its condition with video endostroboscopy (K. Storz) or video pharyngolaryngoscopy (VIVIDEO, KayPentax). Objective acoustic analysis was performed using the Multi-Speech software and hardware complex and the MDVP software (KayPentax). The numerical and graphical expressions of the following parameters were evaluated: noise-to-harmonic ratio (NHR), voice turbulence index (VTI), and soft phonation index (SPI). The results of the study showed that the SPI index is the most informative in assessing hoarseness in patients with hypotonic functional dysphonia. The lower values of this indicator in vocalists can be explained by the presence of a sufficiently developed resonator system of the vocal apparatus, which is reflected in the amplification of the high-frequency components of the voice spectrum.

Machine listening in spatial acoustic scenes with deep networks in different microphone geometries

Multi-channel acoustic source localization evaluates direction-dependentinter-microphone differences in order to estimate the position of an acousticsource embedded in an interfering sound field. We here investigate a deep neuralnetwork (DNN) approach to source localization that improves on previous workwith learned, linear support-vector-machine localizers. DNNs with depthsbetween 4 and 15 layers were trained to predict azimuth direction of targetspeech in 72 directional bins of width 5 degree, embedded in an isotropic,multi-speech-source noise field. Several system parameters were varied, inparticular number of microphones in the bilateral hearing aid scenario wasset to 2, 4, and 6, respectively. Results show that DNNs provide a clear improvement inlocalization performance over a linear classifier reference system.Increasing the number of microphones from 2 to 4 results in a larger increase ofperformance for the DNNs than for the linear system. However, 6 microphonesprovide only a small additional gain. The DNN architectures perform betterwith 4 microphones than the linear approach does with 6 microphones, thusindicating that location-specific information in source-interference scenariosis encoded non-linearly in the sound field.

Surgical Treatment Modalities for Iatrogenic Anterior Glottic Stenosis

Objectives: The aim of this study was to describe common iatrogenic causes of anterior glottic stenosis and to investigate the surgical treatment outcomes of 3 different techniques aimed at repairing stenoses and preventing web reformation. Methods: Clinical information regarding patient demographics, presentation, etiology, treatment, and follow-up outcomes was obtained retrospectively through a review of patients’ records between April 2010 and April 2017. Voice quality was evaluated using grade, roughness, breathiness, asthenia, and strain scale and Voice Handicap Index scores. Objective acoustic parameters were collected using the Kay Elemetrics Multi-Speech program. Results: Twenty patients were included in the study. Causes of glottic web formation included chemoradiation therapy and previous laryngeal surgery for polyps or malignant tumors at the anterior commissure. Web lysis with a CO2 laser, intralaryngeal suturing, and keel placement were used in surgical treatment. Satisfactory results were achieved in 19 patients, and 1 recurrence was observed. Grade, roughness, breathiness, asthenia, and strain scale and Voice Handicap Index scores and objective voice variables for patients with Cohen type I or II webs improved significantly after the surgery. Conclusions: Treatment modalities for anterior glottic stenosis should be carefully chosen according to the etiology of the web, as well as the length and depth of the glottis involved.

Considerações sobre modificações vocais e laríngeas ocasionadas pelo som basal em mulheres sem queixa vocal

O objetivo deste estudo foi verificar as mudanças vocais e laríngeas ocasionadas pelo som basal em cinco indivíduos adultos do sexo feminino sem alterações vocais e laríngeas. Para isso, realizou-se gravação digital da emissão da vogal /a/ e do exame videolaringoestroboscópico. Em seguida, os sujeitos realizaram o som basal durante três séries de 15 repetições, com intervalo de 30 segundos entre cada série, e realizaram-se novamente o exame laríngeo e a gravação da vogal /a/ sustentada. Os dados laríngeos e vocais pré e pós-realização da técnica foram submetidos às análises acústica, perceptivo-auditiva e videolaringoestroboscópica. A análise acústica foi gerada pelo programa Multi Speech. Constatou-se, após o som basal: aumento da vibração da mucosa das pregas vocais; alteração ou manutenção do tipo de voz e do pitch; diminuição ou manutenção das medidas relacionadas ao jitter e shimmer e do índice que sugere ruído glótico; diminuição do índice de fonação suave; manutenção ou alteração da qualidade vocal e do foco ressonantal, com predomínio laringofaríngeo; diminuição da frequência fundamental; e aumento da variação da frequência e da amplitude. Concluiu-se que, nesta sequência de casos, o som basal promoveu efeito positivo sobre a vibração da mucosa das pregas vocais e sobre o ruído da voz, e efeito negativo sobre a ressonância e a estabilidade da voz.

Estudo da constância de medidas acústicas de vogais prolongadas e consecutivas em mulheres sem queixa de voz e em mulheres com disfonia

OBJETIVO: Avaliar a constância de medidas acústicas da frequência fundamental, jitter em porcentagem, quociente de perturbação de frequência (QPF), shimmer em dB, shimmer em porcentagem, quociente de perturbação de amplitude (QPA) e proporção harmônico-ruído (PHR), extraídas de emissões sustentadas e consecutivas de uma mesma vogal, em mulheres sem queixa vocal e em mulheres com disfonia. MÉTODOS: Foram analisados os parâmetros acústicos selecionados em três grupos: 20 mulheres sem queixa vocal, 20 com disfonia e nódulo vocal e dez com disfonia e edema de Reinke, utilizando o programa Multi-Speech Model 3700 da Kay Elemetrics®; utilizou-se a vogal sustentada "é", emitida 15 vezes consecutivas. RESULTADOS: Os valores de frequência fundamental apresentaram variações estatisticamente significantes até a 6ª emissão nos grupos sem queixa vocal e com disfonia e nódulo vocal. Os valores de jitter, QPF, shimmer em dB, shimmer em porcentagem, QPA e PHR não apresentaram diferenças estatisticamente significantes nos três grupos estudados, nas consecutivas emissões. CONCLUSÕES: As medidas dos sinais acústicos, em diferentes emissões consecutivas de uma mesma vogal sustentada são constantes, à exceção da frequência fundamental que pode se deslocar para o agudo nas primeiras emissões de indivíduos sem lesões laríngeas e queixas vocais e com disfonia e nódulos de pregas vocais.