Models included in the AMT (0.9.9)

In order to describe the quality of the models available in the AMT, we rate the implementation of every model by considering its source code and documentation. We also rate the models in terms of their verification, i.e., we rate the results of the implementation versus the results shown in the corresponding publication. The comparison is done within the experiments implemented in the exp_ functions. In the best case, the experiments produce the same results as in the publication - up to some minor layout issues in the graphical representations.

The following table provides an overview of the available models, their documentation, code, and verification status.

Peripheral modelsFunctionDocCodeVerification
Continuous-azimuth HRTF measurementenzner2008
Directional time-of-arrival (on-axis only)ziegelwanger2013
Directional time-of-arrival in HRTFs (off-axis, robust)ziegelwanger2014
Gammatone filterbankgammatone
Invertible Gammatone filterbankhohmann2002
Dual-resonance nonlin. filterbanklopezpoveda1986, 2001
Cochlear transmission-line modelverhulst2012
Auditory-nerve filterbank (humanized cat)zilany2007
Auditory-nerve filterbank (updated synapse model)zilany2014
Auditory brainstem responsesroenne2012
Signal detection modelsFunctionDocCodeVerification
Monaural maskingdau1997
Monaural intensity discrimination, tone-in-noise and AM detection, and spectral and forward masking from Jepsen et al. (2008)jepsen2008_preproc
Binaural signal detectionbreebaart2001
Spatial modelsFunctionDocCodeVerification
Lateralization, cross-correlationlindemann1986
Concurrent-speakers lateral directiondietz2011
Binaural scene analysisspille2013
Lateralization in sound reproduction systemswierstorf2013
Lateralization, supervised trainingmay2011
Binaural activity maptakanen2013
Median-plane localizationlangendijk2002
Localization in cochlear-implant listenerskelvasa2015
Sagittal-plane localization (simple)baumgartner2013
Sagittal-plane localization (robust)baumgartner2014
Sagittal-plane localization (nonlinear, for hearing impairements)baumgartner2016
Sound externalizationbaumgartner2017
Distance perceptiongeorganti2013
Speech perception modelsFunctionDocCodeVerification
Intelligibility in noisejoergensen2011, joergensen2013
Short-time objective intelligibility taal2011
Spatial unmaskingjelfs2011
Data from various publicationsFunctionDocCodeVerification
HRTFs and listener-specific sensitivities from Baumgartner et al. (2013)data_baumgartner2013
HRTFs and listener-specific sensitivities from Baumgartner et al. (2014)data_baumgartner2014
HRTFs and listener-specific sensitivities from Baumgartner et al. (2016)data_baumgartner2016
Localization errors and SCCs from Best et al. (2005)data_best2005
Externalization ratings from Boyd et al. (2012)data_best2012
BMLD thresholds from Breebaart et al. (2001)data_breebaart2001
ABR wave V data from Elberling et al. (2010)data_elberling2010
Notched-noise masking thresholds for the ERB scaledata_glasberg1990
Stapes footplate diplacement from Goode et al. (1994)data_goode1994
Localization performance in sagittal planes from Goupell et al. (2013)data_goupell2013
Tone burst stimuli from Harte et al. (2009)data_harte2009
Externalization ratings from Hartmann and Wittenberg (1996)data_hartmann1996
Externalization ratings from Hassager et al. (2016)data_hassager2016
SRTs tested by Joergensen and Dau (2011)data_joergensen2011
Localization performance and HRTFs from Langendijk et al. (2002)data_langendijk2002
Data from Lindemann (1986a)data_lindenmann1986
Outer and middle ear filter from Lopez-Poveda et al. (2001)data_lopezpoveda2001
Localization polar error rates from Macpherson et al. (2003)data_macpherson2003
Localization performance from Majdak et al. (2010)data_majdak2010
Localization training performance from Majdak et al. (2013)data_majdak2013
Localization performance (CTC condition) from Majdak et al. (2013b)data_majdak2013ctc
Localization performance (non-individualized) from Middlebrooks (1999)data_middlebrooks1999
ABR wave V data as functon of level and sweeping rate from Neely et al. (1988)data_neely1988
Responses to amplitude panning in median plane from Pulkki (2001)data_pulkki2001
"Unity responses" from Roenne (2012)data_roenne2012
Localization response gains from Sabin et al. (2005)data_sabin2005
Data involved in the modeling process of Takanen et al. (2005)data_takanen2005
Masking threshold (binaural/monaural) from van der Par and Kohlrausch (1999)data_vandepar1999
Data involved in the modeling process of Wierstorf et al. (2013)data_wierstorf2013
HRTFs and other data involved in Ziegelwanger et al. (2013)data_ziegelwanger2013
HRTFs and other data involved in Ziegelwanger et al. (2014)data_ziegelwanger2014
Bark scale according to Zwicker (1961)data_zwicker1961
Data from sound externalization experiments Baumgartner et al. (2017)data_baumgartner2017looming
HRTFs and data of a sound externalization model Baumgartner et al. (2017b)data_baumgartner2017
Legend for the documentation and the code:
  • (Submitted): The model has been submitted to the AMT. There is, however, no working code/documentation in the AMT, or there are compilation errors, or some libraries are missing. The current state of the integration can be provided upon request. In the release version, the model neither appears on the website nor is available for download.
  • (OK): The code fits the AMT conventions just enough for being available for download. The model and its documentation appear on the website, but major work is still required.
  • (Good): The code/documentation follows our conventions, but there are open issues.
  • (Perfect): The code/documentation is fully up to our conventions, no open issues.

Legend for the verifications:
  • (Unknown): The AMT can not run experiments for this model and can not produce results for a verification. This might be the case when the verification code has not been provided or nobody has verified yet.
  • (Untrusted): The verification code is available but the experiments do not reproduce the relevant parts of the publication. Such an implementation shall not be trusted as a basis for further developments.
  • (Qualified): The experiments produce similar results as in the publication in terms of showing trends and explaining the effects, but not necessarily matching the numerical results. Explanation for the differences can be provided, for example, not all original data available, or publication affected by a known and documented bug.
  • (Verified): The experiments produce the same results as in the publication. Minor differences are allowed if randomness is involved in the model, for instance, noise as input signal, probabilistic modeling approaches, and a plausible explanation is provided
Note that the description of a model implementation is only a snapshot of the development since the implementations in the AMT are continuously developed, evaluated, and improved.

Model overview of previous releases: