BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CiviCRM//NONSGML CiviEvent iCal//EN X-WR-TIMEZONE:Europe/London METHOD:PUBLISH BEGIN:VEVENT UID:CiviCRM_EventID_732_ba91fa149470137c1611ef676ba3c070@www.ioa.org.uk SUMMARY:London: LSBU Student Projects DESCRIPTION:London Branch Evening Meeting\n \n Wednesday 13 Ju ly 2022\n \n at 6 pm on Zoom\n \n London South Ban k Student Projects\n \n Student Projects:\n \n Imp roving the Intelligibility of Underground Stations using Horizontal Line Arrays\n \n An Audio-Based Vehicle Classifier Using Convolutional Neural Netw ork\n \n Using Non-wave-based Modelling to Explore How much Acoustic Diffusion is too much\n \n Pres entations by Diego Cordes\, Ekim Bakirci and Micha el Fort.\n \n  \;\n \n The students on the Lon don South Bank University Masters course in Enviro nmental and Architectural Acoustics each must unde rtake a dissertation. Presented are three of the b est projects.\n \n The first dissertation investig ates the how to improve speech intelligibility on London Underground by applying modern technology i n the form of Line Arrays. There has been a strugg le to achieve the minimal life safety intelligibil ity requirements on station platforms making milli ons of daily passengers&rsquo\; life difficult. Th rough computer simulations the PA/VA system was ch anged to a planar source to determine how much thi s could improve the intelligibility of the system. By creating a homogeneous sound distribution clos e to the public\, with minimal room excitation thu s minimising reverberations destructive interferen ce to intelligibility could be mitigated resulting in significantly improvements to communication.\n \n The second presentation focuses on audio-based event and scene classification. Many examples of environmental noise detection\, vehicle classifica tion\, and soundscape analysis are developed using state of art deep learning techniques. \; The major noise source in urban and rural areas is ro ad traffic noise. To develop an audio-based vehicl e classifier a convolutional neural network-based algorithm was proposed using Mel spectrogram of au dio signals as an input feature. Different variati ons of the network were generated by changing the parameters of the convolutional layers and the len gth of the network. filter size\, number of filter s were tested with a dataset prepared with various real-life traffic records and audio extracts from traffic videos. The precision of the networks was evaluated with \; common performance metrics. Further assessments were conducted with longer au dio files and predictions of the system compared w ith actual traffic flow.\n \n The final project ex plores the relationship between acoustic surface d iffusion and acoustical character in concert halls by using non-waved based modelling. Measurements were undertaken in a small concert hall to use as a baseline\, which is then transferred into ODEON to experiment with the acoustic parameters further . The relationship between amount of diffusion vs scattering coefficient to achieve optimal objectiv e parameters in a concert hall were generated. The se trends were then tested on existing famous conc ert hall models\, e.g. Musikvereinssaal. The exper iments showed that when considering how much diffu sion is too much in a concert hall\, the statistic al \; amount of diffusion is less of a factor\ , when compared to retaining strong early reflecti ons from side walls and ceilings and the amount of scattering provided by diffuse surfaces.\n \n All three projects and many more will be presented by London South Bank University students in the upco ming Inter-Noise conference to be held in Glasgow\ , August 2022.\n \n For more details of IOA region al branch activities please refer to the IOA websi te: www.ioa.org.uk \;\n \n Topics and speakers for the evening meetings are generally identified and organised by the London Branch Committee\, bu t we always welcome new ideas and suggestions for future presentations. If you have any ideas or sug gestions\, or may even like to give a presentation yourself\, please contact the London Branch Commi ttee Roslyn Andrews. Roslyn.Andrews@aecom.com\n \n This meeting is open to Members and Non Members\n \n The presentations will be recorded if the spea kers give permission.\n \n The Zoom details have a lready been emailed to London Branch members but c an be requested by email at ioa@ioa.org.uk CATEGORIES:Conference CALSCALE:GREGORIAN DTSTAMP;VALUE=DATE-TIME:20220713T180000 DTSTART;VALUE=DATE-TIME:20220713T180000 DTEND;VALUE=DATE-TIME:20220713T180000 URL:https://www.ioa.org.uk/civicrm/event/info?reset=1&id=732 END:VEVENT END:VCALENDAR