Subject: [AUDITORY] Comparison ASH/HSR; ChiME From: Levity <allevity@xxxxxxxx> Date: Mon, 13 Feb 2017 14:44:13 +0000 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>--94eb2c04ec4c2f6ebe05486a7b10 Content-Type: text/plain; charset=UTF-8 Dear List, First, thank you for the useful information often patiently laid down on this mailing list, where no flame war or spam takes place [1]. In the process of writing my PhD dissertation, I am looking for the latest results in Automatic Speech Recognition, compared to Human Speech Recognition. To this end, I have a main question (a) and a side question (b) to you: (a) What are the most recent significant results comparing human and machine speech recognition capabilities? (b) Why has the CHiME challenge stopped providing this comparison [2]? I appreciate the recognition task should be harder every year, but surely this comparison is of interest to the community? Best regards, -- Alban, University of Nottingham [1] http://www.scs.stanford.edu/~dm/home/papers/remove.pdf [2] Results from the CHiME challenges: 1st: http://www.sciencedirect.com/science/article/pii/S0885230812000861 (human SR) 2nd: https://www.merl.com/publications/docs/TR2013-119.pdf (human SR from 1st challenge) 3rd: https://hal.inria.fr/hal-01211376/document (no human SR) 4th: http://spandh.dcs.shef.ac.uk/chime_challenge/results.html (no human SR) --94eb2c04ec4c2f6ebe05486a7b10 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr">Dear List,<div><div><br></div><div>First, thank you for th= e useful information often patiently laid down on this mailing list, where = no flame war or spam takes place [1].=C2=A0</div><div><br></div><div>In the= process of writing my PhD dissertation, I am looking for the latest result= s in Automatic Speech Recognition, compared to Human Speech Recognition. To= this end, I have a main question (a) and a side question (b) to you:=C2=A0= </div><div><br></div><div>(a) What are the most recent significant results = comparing human and machine speech recognition capabilities?=C2=A0</div></d= iv><div><br></div><div>(b) Why has the CHiME challenge stopped providing th= is comparison [2]? I appreciate the recognition task should be harder every= year, but surely this comparison is of interest to the community?</div><di= v><br></div><div>Best regards,=C2=A0<br></div><div>--=C2=A0</div><div>Alban= , University of Nottingham</div><div><br></div><div>[1]=C2=A0<a href=3D"htt= p://www.scs.stanford.edu/~dm/home/papers/remove.pdf">http://www.scs.stanfor= d.edu/~dm/home/papers/remove.pdf</a></div><div>[2] Results from the CHiME c= hallenges:</div><div><div>1st:=C2=A0<a href=3D"http://www.sciencedirect.com= /science/article/pii/S0885230812000861">http://www.sciencedirect.com/scienc= e/article/pii/S0885230812000861</a> (human SR)</div><div>2nd:=C2=A0<a href= =3D"https://www.merl.com/publications/docs/TR2013-119.pdf">https://www.merl= .com/publications/docs/TR2013-119.pdf</a> (human SR from 1st challenge)</di= v><div>3rd:=C2=A0<a href=3D"https://hal.inria.fr/hal-01211376/document">htt= ps://hal.inria.fr/hal-01211376/document</a> (no human SR)</div><div>4th:=C2= =A0<a href=3D"http://spandh.dcs.shef.ac.uk/chime_challenge/results.html">ht= tp://spandh.dcs.shef.ac.uk/chime_challenge/results.html</a> (no human SR)</= div></div></div> --94eb2c04ec4c2f6ebe05486a7b10--