Subject: Re: [AUDITORY] Electric vehicles acoustic discrimination From: LEONARDO GABRIELLI <l.gabrielli@xxxxxxxx> Date: Wed, 3 Jan 2024 10:07:00 +0000--_000_DBBPR04MB62820B5228145DBA42207A9AA960ADBBPR04MB6282eurp_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Dear Frederico, I just want to contribute with a little field experience although I have no= references relevant to your classification problem. Before 2020 I have published works and a patent on the classification of as= phalt roughness and wetness from microphones placed inside a car. The inter= ference of the engine exhaust was low and it was notable only when the mike= (omnidirectional) was placed at 20-40cm from the exhaust pipe, therefore t= he engine sound is nowadays hard to use as a classification hint for deep l= earning algorithms. In recent cars (let's say Euro5 onward) the engine acou= stic emissions are hardly heard at speeds >50 km/h, cause wind and tyre-roa= d noise are predominant. On highways, the presence of a 6th gear or automat= ic will also make the engine spin at relatively low RPM, making the task mo= re difficult, while at low speeds your problem may be easier to tackle. Tak= e care of producing a dataset that involves different floor conditions and = roughness, cause these will greatly affect the sound produced by passing ve= hicles and a good generalization requires involving several tarmac types. Do consider possible biases: it may be easy to discriminate older vehicles = (with higher engine acoustic emissions and/or with high RPM on highways) fr= om recent ones, but more difficult to discriminate EVs vs. internal combust= ion vehicles. Bear this in mind when preparing your dataset and evaluation = because it may bias the result, or conversely, removing relatively silent I= C cars from the dataset may help you gain nice results but bad real-world p= erformance. Finally, I see source localization as a major problem if you do= n't use very directional microphones or beamforming, cause many cars will b= e passing by at toll plazas or highways. For some acoustic information on different terrain types you can give a loo= k at the seminal paper from Abdic et al. related to wetness detection: "Det= ecting Road Surface Wetness from Audio: A Deep Learning Approach". For some more info on the effect of roughness take a look at the papers: "D= eep Neural Networks for Road Surface Roughness Classification from Acoustic= Signals" (144 AES convention), "Processing Acoustic Data with Siamese Neur= al Networks for Enhanced Road Roughness Classification"(IJCNN2019) Good luck with your research work ________________________________ From: AUDITORY - Research in Auditory Perception <AUDITORY@xxxxxxxx>= on behalf of Frederico Pereira <pereira.frederico@xxxxxxxx> Sent: Tuesday, January 2, 2024 7:54 PM To: AUDITORY@xxxxxxxx <AUDITORY@xxxxxxxx> Subject: [AUDITORY] Electric vehicles acoustic discrimination Dear Auditory List, I hope this message finds you well. I am currently in search of published research pertaining to the discrimina= tion of electric vehicles compared to internal combustion vehicles based on= noise emissions. Specifically, I am interested in studies that explore the= feasibility of implementing such discrimination systems in highway enviro= nments (such as installed in span structures, toll plazas, etc..). I acknowledge inherent challenges associated with this task, including issu= es related to extraneous noise, the utilization of Acoustic Vehicle Alertin= g Systems (AVAS), and, notably(?), the dominance of tire-road noise at high= er velocities. My particular interest lies in uncovering research that delv= es into specific acoustic features that may be extracted from microphone ar= ray beamforming systems. From searches on Google Scholar, I was unable to find substantial literatur= e on this specific topic. Therefore, I am reaching out to the esteemed memb= ers of this community to inquire if any of you could direct me towards rele= vant literature addressing this topic. Thank you in advance for your assistance. - Frederico -- Frederico Pereira Mobile:+351 937356301 Audio and Acoustics, Perception Interaction and Usability Email:pereira.frederico@xxxxxxxx<mailto:Email%3Apereira.frederico@xxxxxxxx= m> --_000_DBBPR04MB62820B5228145DBA42207A9AA960ADBBPR04MB6282eurp_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <html> <head> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"= > <style type=3D"text/css" style=3D"display:none;"> P {margin-top:0;margin-bo= ttom:0;} </style> </head> <body dir=3D"ltr"> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> Dear Frederico,</div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> I just want to contribute with a little field experience although I have no= references relevant to your classification problem.</div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <br> </div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> Before 2020 I have published works and a patent on the classification of as= phalt roughness and wetness from microphones placed inside a car. The inter= ference of the engine exhaust was low and it was notable only when the mike= (omnidirectional) was placed at 20-40cm from the exhaust pipe, therefore the engine sound is nowadays hard= to use as a classification hint for deep learning algorithms. In recent ca= rs (let's say Euro5 onward) the engine acoustic emissions are hardly heard = at speeds >50 km/h, cause wind and tyre-road noise are predominant. On highways, the presence of a 6<span><su= p>th</sup> gear or automatic will also make the engine spin at relativ= ely low RPM, making the task more difficult, while at low speeds your probl= em may be easier to tackle. Take care of producing a dataset that involves different floor conditions and roughn= ess, cause these will greatly affect the sound produced by passing vehicles= and a good generalization requires involving several tarmac types.</span><= /div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D"">Do consider possible biases: it may be easy to discriminat= e older vehicles (with higher engine acoustic emissions and/or with high RP= M on highways) from recent ones, but more difficult to discriminate EVs vs.= internal combustion vehicles. Bear this in mind when preparing your dataset and evaluation because it may bia= s the result, or conversely, removing relatively silent IC cars from the da= taset may help you gain nice results but bad real-world performance. Finall= y, I see source localization as a major problem if you don't use very directional microphones or beamformi= ng, cause many cars will be passing by at toll plazas or highways.</span><b= r> </div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D""><br> </span></div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D"" class=3D"ContentPasted2">For some acoustic information on = different terrain types you can give a look at the seminal paper from Abdic= et al. related to wetness detection: "Detecting Road Surface Wetness = from Audio: A Deep Learning Approach".</span></div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D"" class=3D"ContentPasted0 ContentPasted1">For some more info= on the effect of roughness take a look at the papers: "Deep Neural Ne= tworks for Road Surface Roughness Classification from Acoustic Signals"= ; (144 AES convention), "Processing Acoustic Data with Siamese Neural <span style=3D"">Networks for Enhanced Road Rough= ness </span></span><span style=3D"">Classification"(IJCNN2019)</s= pan></div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D""><br> </span></div> <div style=3D"font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, = Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" clas= s=3D"elementToProof"> <span style=3D"">Good luck with your research work</span></div> <div id=3D"appendonsend"></div> <hr style=3D"display:inline-block;width:98%" tabindex=3D"-1"> <div id=3D"divRplyFwdMsg" dir=3D"ltr"><font face=3D"Calibri, sans-serif" st= yle=3D"font-size:11pt" color=3D"#000000"><b>From:</b> AUDITORY - Research i= n Auditory Perception <AUDITORY@xxxxxxxx> on behalf of Frederi= co Pereira <pereira.frederico@xxxxxxxx><br> <b>Sent:</b> Tuesday, January 2, 2024 7:54 PM<br> <b>To:</b> AUDITORY@xxxxxxxx <AUDITORY@xxxxxxxx><br> <b>Subject:</b> [AUDITORY] Electric vehicles acoustic discrimination</font> <div> </div> </div> <div> <div dir=3D"ltr"><span style=3D"color:rgb(33,33,33)">Dear</span><span class= =3D"x_m_9137478987894793065apple-converted-space" style=3D"color:rgb(33,33,= 33)"> </span><span class=3D"x_m_9137478987894793065outlook-search-high= light" style=3D"color:rgb(33,33,33)">Auditory</span><span class=3D"x_m_9137= 478987894793065apple-converted-space" style=3D"color:rgb(33,33,33)"> <= /span><span style=3D"color:rgb(33,33,33)">List,</span> <div><br> </div> <div>I hope this message finds you well. </div> <div>I am currently in search of published research pertaining to the discr= imination of electric vehicles compared to internal combustion vehicles bas= ed on noise emissions. Specifically, I am interested in studies that explor= e the feasibility of implementing such discrimination systems in highway environments (such as install= ed in span structures, toll plazas, etc..).</div> <div><br> </div> <div>I acknowledge inherent challenges associated with this task, including= issues related to extraneous noise, the utilization of Acoustic Vehicle Al= erting Systems (AVAS), and, notably(?), the dominance of tire-road noise at= higher velocities. My particular interest lies in uncovering research that delves into specific acoustic fe= atures that may be extracted from microphone array beamforming systems= .<br> </div> <div><br> </div> <div>From searches on Google Scholar, I was unable to find substantial lite= rature on this specific topic. Therefore, I am reaching out to the esteemed= members of this community to inquire if any of you could direct me towards= relevant literature addressing this topic.</div> <div> <br> </div> <div>Thank you in advance for your assistance.<br> </div> <div><br> </div> <div>- Frederico</div> <div><br> </div> <div>-- <br> </div> <div> <div dir=3D"ltr" class=3D"x_gmail_signature" data-smartmail=3D"gmail_signat= ure"> <div dir=3D"ltr">Frederico Pereira<br> Mobile:+351 937356301 <div><i>Audio and Acoustics, Perception Interaction and Usability</i><br> <div> <div><a href=3D"mailto:Email%3Apereira.frederico@xxxxxxxx" target=3D"_blan= k">Email:pereira.frederico@xxxxxxxx</a></div> </div> </div> </div> </div> </div> </div> </div> </body> </html> --_000_DBBPR04MB62820B5228145DBA42207A9AA960ADBBPR04MB6282eurp_--