Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Paper: A Real-Time 3D Audio Simulator for Cognitive Hearing Science #481

Open
wants to merge 17 commits into
base: 2019
from

Conversation

Projects
None yet
3 participants
@mwickert
Copy link

commented May 23, 2019

If you are creating this PR in order to submit a draft of your paper,
see http://procbuild.scipy.org/ for logs generated by the build
process.

See the project readme
for more information.

@deniederhut deniederhut changed the title Initial pull request Paper: A Real-Time 3D Audio Simulator for Cognitive Hearing Science May 23, 2019

@deniederhut

This comment has been minimized.

Copy link
Member

commented Jun 8, 2019

Hey @mwickert ! There are a lot of sections in your paper that have something like this:

More writing TBD.

Did you forget to push some commits with the new content?

@mwickert

This comment has been minimized.

Copy link
Author

commented Jun 8, 2019

I am working on it now and hope to have more gaps filled later today, and hopefully all completed Sunday evening.

Is this reasonable by you?

@deniederhut

This comment has been minimized.

Copy link
Member

commented Jun 8, 2019

@senolpiskin
Copy link

left a comment

Hey Mark,

I am reviewing this paper for SciPy 2019. After a quick review, I think:

PDF files within the RST file are not viewed automatically.
Some of the links are not functioning correctly.
Youtube video is not available.

Could you please check these issues so that I can review extensively.

Regards,
Senol

@mwickert

This comment has been minimized.

Copy link
Author

commented Jun 10, 2019

Senol,

Working from the bottom up, I would not expect Youtube video to do anything at this point.

When you say links not functioning do you mean in the paper itself?

Finally as for viewing pdf file, I am not sure what else I need to do. I wrote two papers for the 2018 Scipy and I am using the same computer and GitHub configuration as I did in the past. The paper builds to pdf fine on my local machine. I assume that when I push to GitHub the build server should take over and build the pdf of the paper on the server.

Maybe Dillon can comment.

Mark

@deniederhut

This comment has been minimized.

Copy link
Member

commented Jun 11, 2019

Correct - the youtube link is a placeholder for when we have the recorded talk online after the conference.

As for rendering, @senolpiskin you can see the paper here http://procbuild.scipy.org/download/mwickert-2019 on the buildserver

@senolpiskin

This comment has been minimized.

Copy link

commented Jun 11, 2019

Yes, I can see now. Thanks. I will send my comments soon.

@senolpiskin

This comment has been minimized.

Copy link

commented Jun 12, 2019

Here is my review report:

The manuscript describes a 3D audio simulator that can be used in real-time applications for cognitive hearing science. The proposed software can be useful in many aspects of hearing applications. It is within the scope of the SciPy conference and the audience can be highly interested in reading the paper and using the corresponding codes. But, the paper needs several major changes as I have noted below:

The first concern is the language of the manuscript: There are many spelling and grammar mistakes that should be addressed. Furthermore, some statements are not clear. I have mentioned only a few of them although there are many others:

Please rephrase: "To that end this simulator
is integral to planned human subject testing where headphone
testing will be used to blind test subjects for their perception of
where a sound source is located."

Use a comma and rephrase: "To produce a synthesized 3D audio sound field starts with a
geometry."

Not clear and there are two "the": " For a given source location
(x, y,z) = (x1,−x2, x3) pointing at the the origin, we transform
from rectangular coordinates to cylindrical coordinates as shown
in Figure 1."

Please use more formal English: "This transformation is motivated by [Fitzpatrick],
and will be explained more fully in a later section. "
For instance, you could say "This transformation is motivated by [Fitzpatrick],
and explained in detail in section Section Name'."

Do you mean 'mentioned'?: "Note also the
second rectangular coordinate frame mention above is the notation
used by CIPIC."

Not clear, please check again and rephrase: "The 3D audio rendering provided by the simulator developed
in this paper, relies on the 1250 HRIR measurements were taken
using the geometrical configuration shown in Figure 2."

Please remove 'The': "For a source on the 1 m reference sphere, or further away, the there is no ..."

Please rephrase: "When the source is inside is the unit sphere sound parallax ..."

Please check for more. There are many others. I suggest a native English speaker revise the whole text.

Other than the language issues, I have some other suggestions. Please see below:

The abstract is repeated in the introduction section of the main text. That is unnecessary. Please remove that part from the introduction.

Define CIBIC abbreviation before its first usage.

Please define the symbols of the mathematical variables such as azimuth angle, elevation,... in the text at the first usage.

What does it mean? "For a given source location
(x, y,z) = (x1,−x2, x3) pointing at the the origin"
Please explain the equation in more detail. It is not very clear.

Use the same terminology throughout the whole manuscript: "we transform
from rectangular coordinates to cylindrical coordinates" and "convert from cylindrical coordinates
to cartesian"
Use either cartesian or rectangular, not both.

It is not clear what kind of analysis is performed in the section 'Spherical Head Model as a Simple Reference HRIR'. Are you reproducing any previous data here? Or comparing the results with some analytical solutions? If so, what is the error? If not, I suggest performing an error analysis of the results compared to some analytical solution.

I am suggesting re-review after these changes.

@mwickert

This comment has been minimized.

Copy link
Author

commented Jun 12, 2019

Senolpiskin,

Thank you for detailed comments. I appreciate the time you have taken to go through my writing, which I acknowledge was not proofed very well. I fact there are details in the spherical head model section that I feel must be added.

I welcome your re-review. I hope to have a response late tomorrow or early Thursday.

@mwickert

This comment has been minimized.

Copy link
Author

commented Jun 23, 2019

@senolpiskin and @deniederhut
Heavily revised version of the original paper posted incrementally over the last 1+ week, now ready for a second review.

In the below I have interleaved by response to your original review.

Here is my review report: The manuscript describes a 3D audio simulator that can be used in real-time applications for cognitive hearing science. The proposed software can be useful in many aspects of hearing applications. It is within the scope of the SciPy conference and the audience can be highly interested in reading the paper and using the corresponding codes. But, the paper needs several major changes as I have noted below:

General Comment: The first concern is the language of the manuscript: There are many spelling and grammar mistakes that should be addressed. Furthermore, some statements are not clear. I have mentioned only a few of them although there are many others:

  • Agree, I have made many edits

Specific Comments: Please rephrase: "To that end this simulator
is integral to planned human subject testing where headphone
testing will be used to blind test subjects for their perception of
where a sound source is located."

  • Done

Use a comma and rephrase: "To produce a synthesized 3D audio sound field starts with a
geometry."

  • Done; sentences reworked and the corresponding figure was reworked to make the two coordinate frames clear. Note the first frame is

Not clear and there are two "the": " For a given source location
(x, y,z) = (x1,−x2, x3) pointing at the the origin, we transform
from rectangular coordinates to cylindrical coordinates as shown
in Figure 1."

  • Done in the rework that also resolved ; What you reference above was indeed not clear, and when I went back to re-read things I concluded it was not correct

Please use more formal English: "This transformation is motivated by [Fitzpatrick],
and will be explained more fully in a later section. "
For instance, you could say "This transformation is motivated by [Fitzpatrick],
and explained in detail in section Section Name'."

  • Fixed

Do you mean 'mentioned'?: "Note also the
second rectangular coordinate frame mention above is the notation
used by CIPIC."

  • Resolved; I likely had a typo here, but in my rewrite I changed the phrase to "The second system is referred to by CIPIC..."

Not clear, please check again and rephrase: "The 3D audio rendering provided by the simulator developed
in this paper, relies on the 1250 HRIR measurements were taken
using the geometrical configuration shown in Figure 2."

  • Fixed; lots of rewrite in this area

Please remove 'The': "For a source on the 1 m reference sphere, or further away, the there is no ..."

  • Fixed; This was an incorrect statement and is now stated correctly in this section of the paper

Please rephrase: "When the source is inside is the unit sphere sound parallax ..."

  • Fixed/resolved; again this section of the paper has undergone a lot of rewrite. I have also now added Figure 6 to give a pictoral view of the parallax problem. There is not enough space to go through all of the equations behind the parallax correction, but the paper by Fitzpatrick is readily available.

Please check for more. There are many others. I suggest a native English speaker revise the whole text.

  • Resolved, I think; The first draft was written too quickly during a crunch time.

Other than the language issues, I have some other suggestions. Please see below:

The abstract is repeated in the introduction section of the main text. That is unnecessary. Please remove that part from the introduction.

  • Duplication removed; this was carelessness on my part

Define CIBIC abbreviation before its first usage.

  • Fixed; See the introduction. I though I had done this, but good that you caught it.

Please define the symbols of the mathematical variables such as azimuth angle, elevation,... in the text at the first usage.

  • In the rewrite I have tried to make all variables clear and defined. Figure 1 for example now sets the stage for a large number of geometry related variables. I refer back to this figure later in the paper to help provide a visual as to how variables fit into the big picture. In the spherical head section some new variables get defined. The greek letter \theta gets used in multiple contexts, but the subscript is unique to the usage (I hope).

What does it mean? "For a given source location
(x, y,z) = (x1,−x2, x3) pointing at the the origin"
Please explain the equation in more detail. It is not very clear.

  • Resolved; This was a confusing section. It has been rewritten and this equality no longer appears.

Use the same terminology throughout the whole manuscript: "we transform
from rectangular coordinates to cylindrical coordinates" and "convert from cylindrical coordinates
to cartesian"
Use either cartesian or rectangular, not both.

  • Fixed; Cartesian is the dominant coordinate system employed, but CIPIC uses IPCS (really a spherical system, I have to interface with this) but now I try hard to make it clear that cylindrical coordinates are only used in the Jupyter notebook apps as a convenience for the user to position the audio source in what I think is a convenient way access localization when informally testing subjects.

It is not clear what kind of analysis is performed in the section 'Spherical Head Model as a Simple Reference HRIR'. Are you reproducing any previous data here? Or comparing the results with some analytical solutions? If so, what is the error? If not, I suggest performing an error analysis of the results compared to some analytical solution.

  • I am going to try to answer your questions first, then move on to justifying my approach. The references that I have provided in the spherical head model section of the paper are classics in my opinion. There are no closed-form analytical solutions for the geometries I am most interested in, but the paper by Duda and Martens has many plots and in particular my Figure 11 is very similar (not the same geometries) their Fig 6. This paper also has pseudo-code in their Appendix B, which is my function HRTF_sph() in the paper. I feel comfortable with HRIR results I have obtained. The Duda and Martens paper does provided every step they took from HRTF to HRIR, so some small differences are possible. At the end of the spherical head section I now include some comments on a listening test that was conducted.

Wrap-up: I am suggesting re-review after these changes.

  • I am ready to move forward with the next round.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.