Re: [AUDITORY] cleaning background noise (murmur, clatter, ...) (Esra Mungan )

Subject: Re: [AUDITORY] cleaning background noise (murmur, clatter, ...) From: Esra Mungan <esra.mungan@xxxxxxxx> Date: Sat, 12 Apr 2025 15:01:17 +0300 This is a multipart message in MIME format. ------=_NextPart_000_002E_01DBABBB.BE00D040 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable dear jan-willem, =20 thank you for your suggestion, we=92ll check out =93descript=94. the = speeches are in turkish, and as said, the least we want is to change the timbre of = the original voice. this happened when we isolated the audio file and had = it then cleaned off of its background noise. it seems that the cleaning = took out too many, say, overtone frequencies (and speaking of frequencies, I = sure will read adam weisser=92s article!), which =97while creating indeed a = very clean&clear audio file (to be re-knitted to the video file)=97 = unfortunately changed the timbre of the speaker=92s voice far too much. =20 I also thank massimo grassi for his suggestion to change the way we = record the sound, i.e, to use earplugs with mics to record from. and the more feedback I am getting the more I=92ve become convinced that the cheapest = and most effective way to go is to record directly from a lavalier mic or bluetooth earplugs.=20 =20 best, esra =20 Esra Mungan, PhD Associate Professor (left Bogazici University since Dec. 2024) https://psychology.bogazici.edu.tr/content/esra-mungan=20 https://universitybogazici.wordpress.com/=20 https://direnenakademi.wordpress.com/ =20 =20 From: Wasmann, Jan-Willem [mailto:Jan-Willem.Wasmann@xxxxxxxx=20 Sent: Friday, April 11, 2025 10:21 AM To: 'Esra Mungan' <esra.mungan@xxxxxxxx>; = AUDITORY@xxxxxxxx Subject: RE: [AUDITORY] cleaning background noise (murmur, clatter, ...) =20 Hi Esra, A very easy tool for editing recordings containing speech is descript. = It will read in the recording and use speech to text to create a = transcription. You can then edit the transcript as if you are working in a word editor. Removing a line will remove the audio and video of that portion. In addition, you can use filters (noise canceling etc) to improve sound quality. I have used this a year ago for some podcast recordings and it worked well. Not sure how the software developed in the meantime. I = don=92t have the paid version anymore. You pay for the file size so video = editing will be much more expensive than audio-only editing. In what language = did you record? It doesn=92t support all languages yet. Also my accent in = English introduced many transcription errors. =20 =20 <https://www.descript.com/podcasting?utm_source=3Dgoogle&utm_medium=3Dpai= d&utm_c ampaign=3D16259794211&utm_campaignname=3DGoogle_Search_Brand_T1&utm_conte= nt=3D1638 44740293&utm_term=3Ddescript%20audio%20tools&utm_kwid=3Dkwd-2317408635568= &gclid=3D EAIaIQobChMI-fLwubfPjAMV7LKDBx1nnAc6EAAYASAAEgKg4PD_BwE&utm_ad_id=3D70569= 05677 61&utm_campaigntype=3Dpaid_search&gad_source=3D1> Descript: AI-Powered = Podcast & Audio Editor =20 All my best, Jan-Willem=20 =20 Van: AUDITORY - Research in Auditory Perception = <AUDITORY@xxxxxxxx <mailto:AUDITORY@xxxxxxxx> > Namens Esra Mungan Verzonden: donderdag 10 april 2025 13:33 Aan: AUDITORY@xxxxxxxx <mailto:AUDITORY@xxxxxxxx>=20 Onderwerp: [AUDITORY] cleaning background noise (murmur, clatter, ...) =20 dear list, =20 since more than a year we are doing a series of weekly public science lectures (as part of a collective resistance against a political crash attempt against our university, hence we call our initiative =93academy-in-resistance for free, autonomous, and democratic = universities=94) at a central public library in istanbul.=20 we have wonderful attendance, yet, there are, of course, many people who cannot attend these lectures in person. =20 all lectures are recorded but because the sound and recording conditions = are not ideal (particularly on warm days we need to keep windows open hence = we get some background city noise; but also our recording devices and the = whole bluetooth sound system are rather modest)=20 we now want to improve the listening quality of our recorded lectures ( <https://www.youtube.com/@xxxxxxxx> https://www.youtube.com/@xxxxxxxx; last years=92 lectures were = added under a more general academic youtube account of ours, <https://www.youtube.com/@xxxxxxxx> https://www.youtube.com/@xxxxxxxx). =20 a colleague brought these programs to our attention, * https://flixier.com/tools/audio-enhancer * https://www.kapwing.com/tools/audio-editor/audio-enhancer * https://www.geeksforgeeks.org/top-free-video-audio-enhancer-tools/ * https://www.canva.com/features/audio-enhancer/ * https://voice.ai/tools/audio-enhancer =20 however, I understand that these can only be used once the videos are uploaded. what we need though is a way to do the cleaning beforehand = (each video recording is about 17-35GB in size). another supporter offered to = do it on adobe premiere and the results were indeed great in terms of = speaker clarity, however this time, the timbre of the speaker had quite changed. = :) some of them are well-known scholars, hence such major changes in voice character feels really alienating. so we are stuck now. =20 is there anything you can suggest us? many thanks, esra =20 =20 Esra Mungan, PhD Associate Professor (left Bogazici University since Dec. 2024) <https://psychology.bogazici.edu.tr/content/esra-mungan> https://psychology.bogazici.edu.tr/content/esra-mungan (TK/EN) <https://universitybogazici.wordpress.com/> https://universitybogazici.wordpress.com/ (TK/EN) <https://direnenakademi.wordpress.com/> https://direnenakademi.wordpress.com/ (TK)=20 =20 De informatie in dit bericht is uitsluitend bestemd voor de = geadresseerde. Aan dit bericht en de bijlagen kunnen geen rechten worden ontleend. = Heeft u deze e-mail onbedoeld ontvangen? Dan verzoeken wij u het te vernietigen = en de afzender te informeren. Openbaar maken, kopi=EBren en verspreiden van = deze e-mail of informatie uit deze e-mail is alleen toegestaan met = voorafgaande schriftelijke toestemming van de afzender. Het Radboudumc staat geregistreerd bij de Kamer van Koophandel in het handelsregister onder nummer 80262783. The content of this message is intended solely for the addressee. No = rights can be derived from this message or its attachments. If you are not the intended recipient, we kindly request you to delete the message and = inform the sender. It is strictly prohibited to disclose, copy or distribute = this email or the information inside it, without a written consent from the sender. Radboud university medical center is registered with the Dutch Chamber of Commerce trade register with number 80262783.=20 ------=_NextPart_000_002E_01DBABBB.BE00D040 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <html xmlns:v=3D"urn:schemas-microsoft-com:vml" = xmlns:o=3D"urn:schemas-microsoft-com:office:office" = xmlns:w=3D"urn:schemas-microsoft-com:office:word" = xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" = xmlns=3D"http://www.w3.org/TR/REC-html40"><head><meta = http-equiv=3DContent-Type content=3D"text/html; = charset=3Diso-8859-1"><meta name=3DGenerator content=3D"Microsoft Word = 15 (filtered medium)"><style></style></head><body lang=3DEN-US = link=3D"#0563C1" vlink=3D"#954F72"><div class=3DWordSection1>dear = jan-willem,<o:p></o:p><o:p> </o:p></p= >thank you for your suggestion, = we’ll check out “descript”. the speeches are in = turkish, and as said, the least we want is to change the timbre of the = original voice.=A0 this happened when we isolated the audio file and had = it then cleaned off of its background noise.=A0 it seems that the = cleaning took out too many, say, overtone frequencies = (and speaking of = frequencies, I sure will read adam weisser’s = article!), which = —while creating indeed a very clean&clear audio file (to be = re-knitted to the video file)— unfortunately changed the timbre of = the speaker’s voice far too much.=A0 = <o:p></o:p>I also thank massimo grassi for = his suggestion to change the way we record the sound, i.e, to use = earplugs with mics to record from.=A0 and the more feedback I am getting = the more I’ve become convinced that the cheapest and most = effective way to go is to record directly from a lavalier mic or = bluetooth earplugs. <o:p></o:p><o:p> </o:p></p= >best,<o:p></o:p>= esra<o:p></o:p><= p class=3DMsoNormal><o:p> </o:p></p= ><div>Esra Mungan, = PhD<o:p></o:p>Associate = Professor<o:p></o:p>(left Bogazici University since = Dec. 2024)<o:p></o:p><a = href=3D"https://psychology.bogazici.edu.tr/content/esra-mungan">https://p= sychology.bogazici.edu.tr/content/esra-mungan</a> = <o:p></o:p><a = href=3D"https://universitybogazici.wordpress.com/">https://universityboga= zici.wordpress.com/</a> <o:p></o:p><a = href=3D"https://direnenakademi.wordpress.com/">https://direnenakademi.wor= dpress.com/</a> =A0<o:p></o:p></div><o:p> </o:p></p= ><div><div style=3D'border:none;border-top:solid #E1E1E1 = 1.0pt;padding:3.0pt 0in 0in 0in'>From: = Wasmann, Jan-Willem [mailto:Jan-Willem.Wasmann@xxxxxxxx = Sent: Friday, April = 11, 2025 10:21 AM To: = 'Esra Mungan' <esra.mungan@xxxxxxxx>; = AUDITORY@xxxxxxxx Subject: RE: [AUDITORY] cleaning = background noise (murmur, clatter, ...)<o:p></o:p></div></div><o:p> </o:p>Hi Esra,<o:p></o:p>A very easy tool for editing recordings = containing speech is descript. It will read in the recording and use = speech to text to create a transcription. You can then edit the = transcript as if you are working in a word editor. Removing a line will = remove the audio and video of that portion. In addition, you can use = filters (noise canceling etc) to improve sound quality. I have used this = a year ago for some podcast recordings and it worked well. Not sure how = the software developed in the meantime. I don’t have the paid = version anymore. You pay for the file size so video editing will be much = more expensive than audio-only editing. In what language did you record? = It doesn’t support all languages yet. Also my accent in English = introduced many transcription errors.<o:p></o:p><o:p> </o:p><a = href=3D"https://www.descript.com/podcasting?utm_source=3Dgoogle&utm_m= edium=3Dpaid&utm_campaign=3D16259794211&utm_campaignname=3DGoogle= _Search_Brand_T1&utm_content=3D163844740293&utm_term=3Ddescript%2= 0audio%20tools&utm_kwid=3Dkwd-2317408635568&gclid=3DEAIaIQobChMI-= fLwubfPjAMV7LKDBx1nnAc6EAAYASAAEgKg4PD_BwE&utm_ad_id=3D705690567761&a= mp;utm_campaigntype=3Dpaid_search&gad_source=3D1">Descript: AI-Powered Podcast & Audio = Editor</a><o:p></o:p><o:p> </o:p>All my best,<o:p></o:p>Jan-Willem <o:p></o:p><o:p> </o:p><div><div = style=3D'border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in = 0in 0in'>Van: AUDITORY - Research in Auditory Perception <<a = href=3D"mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a>>= Namens Esra = Mungan Verzonden: = donderdag 10 april 2025 13:33 Aan: <a = href=3D"mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a> = Onderwerp: [AUDITORY] = cleaning background noise (murmur, clatter, = ...)<o:p></o:p></div></div><o:p> </o:p>dear = list,<o:p></o:p><o:p> </o:p></p= >since more than a year we are = doing a series of weekly public science lectures (as part of a = collective resistance against a political crash attempt against our = university, hence we call our initiative “academy-in-resistance = for free, autonomous, and democratic = universities”) = at a central public library in = istanbul. <o:p></o:p>we have wonderful attendance, = yet, there are, of course, many people who cannot attend these lectures = in person.  <o:p></o:p>all lectures are recorded but = because the sound and recording conditions are not ideal = (particularly = on warm days we need to keep windows open hence we get some background = city noise; but also our recording devices and the whole bluetooth sound = system are rather modest) = <o:p></o:p>we now want to improve the = listening quality of our recorded lectures (<a = href=3D"https://www.youtube.com/@xxxxxxxx">https://www.youtube.com/@xxxxxxxx</f= ont></a>; last years’ lectures were added under a more general = academic youtube account of ours, <a = href=3D"https://www.youtube.com/@xxxxxxxx">https://www.youtube.com/@xxxxxxxx</spa= n></a>).<o:p></o:p><o:p> </o:p></p= >a colleague brought these = programs to our attention,<o:p></o:p><ul = style=3D'margin-top:0in' type=3Ddisc><li class=3DMsoNormal = style=3D'color:black;mso-list:l1 level1 lfo3'><a = href=3D"https://flixier.com/tools/audio-enhancer" = target=3D"_blank">https://flixier.com/tools/audio-enhancer</a><o:p></o:p>= </li><li class=3DMsoNormal = style=3D'color:black;mso-list:l1 level1 lfo3'><a = href=3D"https://www.kapwing.com/tools/audio-editor/audio-enhancer" = target=3D"_blank">https://www.kapwing.com/tools/audio-editor/audio-enhanc= er</a><o:p></o:p></li><li class=3DMsoNormal = style=3D'color:black;mso-list:l1 level1 lfo3'><a = href=3D"https://www.geeksforgeeks.org/top-free-video-audio-enhancer-tools= /" = target=3D"_blank">https://www.geeksforgeeks.org/top-free-video-audio-enha= ncer-tools/</a><o:p></o:p></li><li class=3DMsoNormal = style=3D'color:black;mso-list:l1 level1 lfo3'><a = href=3D"https://www.canva.com/features/audio-enhancer/" = target=3D"_blank">https://www.canva.com/features/audio-enhancer/</a><o:p>= </o:p></li><li class=3DMsoNormal = style=3D'color:black;mso-list:l1 level1 lfo3'><a = href=3D"https://voice.ai/tools/audio-enhancer" = target=3D"_blank">https://voice.ai/tools/audio-enhancer</a><o:p></o:p></s= pan></li></ul><o:p> </o:p></p= >however, I understand that these = can only be used once the videos are uploaded. what we need though is = a way to do the cleaning = beforehand (each video recording is about 17-35GB in size). =  another supporter offered to do it on adobe premiere and the = results were indeed great in terms of speaker clarity, however this = time, the timbre of the speaker had quite changed. :)  some of them = are well-known scholars, hence such major changes in voice character = feels really alienating.  so we are stuck = now.<o:p></o:p><o:p> </o:p></p= >is there anything you can suggest = us?<o:p></o:p>many = thanks,<o:p></o:p>esra<o:p></o:p><= p class=3DMsoNormal><o:p> </o:p></p= ><o:p> </o:p></p= >Esra Mungan, = PhD<o:p></o:p>Associate = Professor<o:p></o:p>(left Bogazici University since = Dec. 2024)<o:p></o:p><a = href=3D"https://psychology.bogazici.edu.tr/content/esra-mungan">https://psychology.bogazici.edu.tr/content/esra-mungan</= a> = (TK/EN)<o:p></o:p><a = href=3D"https://universitybogazici.wordpress.com/">https://universitybogazici.wordpress.com/</a> = (TK/EN)<o:p></o:p></p= ><a = href=3D"https://direnenakademi.wordpress.com/">https://direnenakademi.wordpress.com/</a> = (TK) = <o:p></o:p><o:p> </o:p><font= size=3D2 face=3DArial>De informatie in dit bericht is uitsluitend bestemd voor de = geadresseerde. Aan dit bericht en de bijlagen kunnen geen rechten worden = ontleend. Heeft u deze e-mail onbedoeld ontvangen? Dan verzoeken wij u = het te vernietigen en de afzender te informeren. Openbaar maken, = kopi=EBren en verspreiden van deze e-mail of informatie uit deze e-mail = is alleen toegestaan met voorafgaande schriftelijke toestemming van de = afzender. Het Radboudumc staat geregistreerd bij de Kamer van Koophandel = in het handelsregister onder nummer 80262783. The content of this = message is intended solely for the addressee. No rights can be derived = from this message or its attachments. If you are not the intended = recipient, we kindly request you to delete the message and inform the = sender. It is strictly prohibited to disclose, copy or distribute this = email or the information inside it, without a written consent from the = sender. Radboud university medical center is registered with the Dutch = Chamber of Commerce trade register with number 80262783. = <o:p></o:p></div></body></html> ------=_NextPart_000_002E_01DBABBB.BE00D040--

This message came from the mail archive
postings/2025/
maintained by:

DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University