News feed

Log in to your course web

You are not logged in KTH, so we cannot customize the content.

Do you have questions regarding this course?

If you are registered on a current course round, please refer to the Canvas course room. You find the right course room under "Courses" in the personal menu.

If you are not registered, see Course PM for DT2112 or contact your student office or study counsellor.

In the News feed, you find updates for pages, schedule and posts from teachers (when aimed also at earlier registered students).

February 2016

Hi all,

I got a question form a student on how to answer the questions about the analysis windows in Lab1. I answer here so that everybody can see.

As we discussed during the lecture, feature frames are computed over short windows of speech samples. The window is then shifted in time to compute the next feature vector and the shift may be smaller than the window length (so that contiguous windows overlap in time).

The number of speech samples per second is given by the sampling frequency. The length of the window and its shift are given by the corresponding configuration parameters. You should be able to relate these three quantities to determine how many samples each window contains and how many milliseconds (or if you wish samples) the windows overlap.

To answer the question about how many feature vectors you have in a typical utterance, you should first decide how long is a typical utterance (you can have a look with wavesurfer at the ones you have created). Then you have to determine which parameter is relevant in this case among the ones mentioned above. You can safely disregard boundary effects at the beginning or end of the utterance because we are talking about a approximate number of feature vectors.

Hope this clarifies the questions

Report abuse

Teacher Giampiero Salvi posted 2 February 2016

Teacher

Giampiero Salvi commented 2 February 2016

Another question is about the topology of the HMM.

Here I just want you to draw a graph indicating the states of the HMM and which transitions between them are allowed according to the transition matrix in the prototype file.

January 2016

Hi,

I am writing this as an example on how to write comments that are visible without login. You choose "The whole world" in the "Can read:" field. This could be a way to solve the problem with the students from Uppsala that have no access to KTH social.

Report abuse

Teacher Giampiero Salvi posted 28 January 2016

Teacher

Jens Edlund commented 29 January 2016

I assume this worked? Can someone from Uppsla cofirm that? Cheers.

Teacher

Giampiero Salvi commented 29 January 2016

Yes, I believe they can see these comments. They still cannot see the other pages and for those there is no such option.

Hi all,

I saw that many are willing to run the lab on their own Ubuntu computers. Here are building instructions for HTK (version 3.4.1) that worked for me on Ubuntu 14.04:

sudo apt-get install libc6-dev-i386
./configure --disable-hslab --disable-hlmtools --prefix=/opt/htk
make all
sudo mkdir /opt/htk
sudo make install

Then you will need to add the path to your PATH variable, for example with adding this line

in bash:

export PATH=/opt/htk/bin:$PATH

in tcsh:

setenv PATH=/opt/htk/bin:$PATH

Test if it works by running any of the HTK commands, for example HList

For the lab to work, you will also need to install Wavesurfer with:

apt-get install libsnack-alsa wavesurfer

Finally, if you do the lab on your computer, disregard the command "source config/environment" from the lab instructions

Good luck!

Report abuse

Teacher Giampiero Salvi posted 28 January 2016

Teacher Giampiero Salvi edited 28 January 2016

Hi all,

I saw that many are willing to run the lab on their own Ubuntu computers. Here are building instructions that worked for me on Ubuntu 14.04:

sudo apt-get install libc6-dev-i386./configure --disable-hslab --disable-hlmtools --prefix=/opt/htkmake allsudo mkdir /opt/htksudo make install

Then you will need to add the path to your PATH variable, for example with adding this line

in bash:

export PATH=/opt/htk/bin:$PATH

in tcsh:

setenv PATH=/opt/htk/bin:$PATH

Test if it works by running any of the HTK commands, for example HList

For the lab to work, you will also need to install Wavesurfer with:¶

apt-get install libsnack-alsa wavesurfer¶

Good luck!¶

Teacher Giampiero Salvi edited 28 January 2016

Hi all,

I saw that many are willing to run the lab on their own Ubuntu computers. Here are building instructions that worked for me on Ubuntu 14.04:

sudo apt-get install libc6-dev-i386./configure --disable-hslab --disable-hlmtools --prefix=/opt/htkmake allsudo mkdir /opt/htksudo make install

Then you will need to add the path to your PATH variable, for example with adding this line

in bash:

export PATH=/opt/htk/bin:$PATH

in tcsh:

setenv PATH=/opt/htk/bin:$PATH

Test if it works by running any of the HTK commands, for example HList

For the lab to work, you will also need to install Wavesurfer with:

apt-get install libsnack-alsa wavesurfer

Finally, if you do the lab on your computer, disregard the command "source config/environment" from the lab instructions¶

¶

Good luck!

Teacher Giampiero Salvi edited 28 January 2016

Hi all,

I saw that many are willing to run the lab on their own Ubuntu computers. Here are building instructions for HTK (version 3.4.1) that worked for me on Ubuntu 14.04:

sudo apt-get install libc6-dev-i386./configure --disable-hslab --disable-hlmtools --prefix=/opt/htkmake allsudo mkdir /opt/htksudo make install

Then you will need to add the path to your PATH variable, for example with adding this line

in bash:

export PATH=/opt/htk/bin:$PATH

in tcsh:

setenv PATH=/opt/htk/bin:$PATH

Test if it works by running any of the HTK commands, for example HList

For the lab to work, you will also need to install Wavesurfer with:

apt-get install libsnack-alsa wavesurfer

Finally, if you do the lab on your computer, disregard the command "source config/environment" from the lab instructions

Good luck!

Teacher

Giampiero Salvi commented 28 January 2016

Hi again,

now I had the time to actually test the commands with a fresh Ubuntu 14.04.3 installation and I realized the above instructions are somewhat incomplete. If you want to run on your own machine I recommend you use the new lab files I made available under the lab1 page (dated 2016-01-28).