Historique: Two-channel mixtures of speech and real-world background noise
Comparaison de la version 86 à la version 89
|
@@ -Lignes: 3-6 changées en +Lignes: 3-11 @@
This task aims at evaluating source separation and denoising techniques in the context of speech enhancement by merging two datasets: the [http://sisec2010.wiki.irisa.fr/tiki-index.php?page=Source+separation+in+the+presence+of+real-world+background+noise|SiSEC 2010 noisy speech dataset] and the [http://www.dcs.shef.ac.uk/spandh/chime/PCC/datasets.html|CHiME corpus]. Both datasets consist of two-channel mixtures of one speech source and real-world background noise, so that algorithms applicable to one dataset are applicable to the other without additional effort. The source separation results obtained over the latter dataset will be analyzed in line of the speech recognition results obtained over that dataset as part of the [http://www.dcs.shef.ac.uk/spandh/chime/challenge.html|CHiME Challenge].
+
+ + !!Results + + __See the results over [http://www.irisa.fr/metiss/SiSEC11/noise/results_test.html|test] and [http://www.irisa.fr/metiss/SiSEC11/noise/results_dev.html|development] data__ @@ -Lignes: 62-65 changées en +Lignes: 67-72 @@
* -+dev_<env>_<cond>_<take>_DOA.txt+-: DOA of the speech source (see the [http://sisec2010.wiki.irisa.fr/tiki-index.php?page=Source+separation+in+the+presence+of+real-world+background+noise|SiSEC 2010 wiki] for the convention adopted to measure DOA)
Since the source DOAs were measured geometrically in the -+Su+- and -+Ca+- environments, they might contain a measurement error up to a few degrees; on the contrary, there is no such error in the -+Sq+- environment.
+
+ The mixtures dev_Ca1_Co_A_mix.wav and dev_Ca1_Co_B_mix.wav are identical (this is a mistake that will be corrected in future evaluations). Entrants wishing to exploit the context of each sentence in the domestic environment database can also __download the corresponding [http://www.irisa.fr/metiss/SiSEC11/noise/dev_embedded.zip|5 min recordings] (86 MB)__ (same nomenclature as above). @@ -Lignes: 117-121 changées en +Lignes: 124-128 @@
The estimated speaker DOAs in task 1 will be evaluated in terms of absolute difference with the true DOAs.
- The estimated speech signals in task 2 will be evaluated via the energy ratio criteria defined in the [http://bass-db.gforge.inria.fr/bss_eval/|BSS_EVAL] toolbox allowing arbitrary filtering between the estimated source and the true source and via the perceptually-motivated criteria in the [http://bass-db.gforge.inria.fr/peass/PEASS-Software.html|PEASS] toolkit.
+ The estimated speech signals in task 2 will be evaluated via the energy ratio criteria defined in the [http://bass-db.gforge.inria.fr/bss_eval/|BSS_EVAL] toolbox allowing arbitrary filtering between the estimated source and the true source.
The estimated speech and noise spatial image signals in task 3 will be evaluated via the energy ratio criteria introduced for the [http://www.irisa.fr/metiss/SASSEC07/?show=criteria|Stereo Audio Source Separation Evaluation Campaign] and via the perceptually-motivated criteria in the [http://bass-db.gforge.inria.fr/peass/PEASS-Software.html|PEASS] toolkit. |