Password Athentication through voice in C#

Question

1.00/5 (2 votes)

See more:

Hai all,

I am devolping an application, for that i need to use voice athentication in C#..

I have saved my password(as voice not text) in SQL server as image datatype,

and while in time of authentication(login).

I am getting user input as .wav file to my Application Startup path.

and i need to check this user input with orginal password which is saved in database

for that i used byte array verfication method, and i used this code for that

C#

            try
            {
                byte[] b1 = File.ReadAllBytes(Application.StartupPath + "\\password.wav");\\ where password.wav is the user input
                byte[] b2 = Getbytes(username);\\this method will return a byte array of password of that corresponding username


\\Checking

                if (b1.SequenceEqual<byte>(b2) == true)
                {
                    MessageBox.Show("Welcome "+username+", "Message", MessageBoxButtons.OK, MessageBoxIcon.Information);
                }
                else
                {
                    MessageBox.Show("Sorry Try again", "Message", MessageBoxButtons.OK, MessageBoxIcon.Exclamation);
                }
            }
            catch (FileNotFoundException ex)
            {
                MessageBox.Show("File Not Found", "Message", MessageBoxButtons.OK, MessageBoxIcon.Warning);
            }

but when tested this code...

the byte array is not getting equal,so that

I can't achieve my goal...

My Qustion is.. Is there any other Idea(method) to achive my goal..

Please help..

Posted 26-Jul-11 20:27pm

version_2.0

Add a Solution

3 solutions

Solution 3

In addition to what is given as solutions, Microsoft provided easy and interesting Speech class in .NET framework 4. Here is the whole MSDN article.System.Speech.Recognition Namespace[^]

Posted 30-Jul-11 11:46am

Wonde Tadesse

Updated 30-Jul-11 11:52am

v2

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

OriginalGriff · Accepted Answer · 2011-07-26T20:33:00

Solution 1

For goodness sake! You need to do some basic research into voice regocnition. As I suggested to you yesterday, google can help, but you are going to have to do some work.

The simple approach you are trying to use will never work. Why not? A first, simple reason:

When did you start talking in each sample? At exactly the same time in microseconds? Because if you didn't there is no way the two samples will match.

A second simple reason:

When you sign your name, are the signatures absolutely identical? Only in CSI... What makes you think your voice will be any different?

Posted 26-Jul-11 20:33pm

OriginalGriff

Comments

version_2.0 27-Jul-11 2:42am

while I am searching google i am getting losts of methods about it

but now, what i need is, Is there any common method(usally doing methods)
to achieve my goal...

I have already searched this System.Speech.SpeechRecognitionEngine, System.Speech.SpeechRecognizer

If possible redirect me to any good tutorial...

Thanks in advance...

OriginalGriff 27-Jul-11 2:49am

No, there isn't. It is a complex subject, not being worked on my that many people. You are going to have to do basic research into what makes a voice sample unique to a person, rather than "where can I find copy'n'paste code from" research.

version_2.0 27-Jul-11 2:55am

ok.. got it.. Thanks..

version_2.0 19-Aug-11 1:28am

First of all sorry,
u were right.
I have to do research on voice.
I took lots of time to realize that. but, I cant do this with out your help.. sir could you please tell me that form where i have to start my research. I am totally blank..

OriginalGriff 19-Aug-11 3:48am

Google, as always: Start here, and follow what looks like it may help. I think you have a big task on your hands though!
http://www.google.co.uk/search?sourceid=chrome&ie=UTF-8&q=basics+of+human+voice+print+identification

version_2.0 27-Jul-11 5:11am

I have a doubt.. Delete the above Question & situation.. I know that with out Grammar System.Speech.Recognition.SpeechRecognizer will not recognize any word(will give error).. How can I recognize a word without language... what i mean was i need the exact text(string) or possible text(string) form the speech.. Is it possible... If possible, I need your help...

Sergey Alexandrovich Kryukov 28-Jul-11 2:42am

I addressed this problem in my solution. You can ask your follow-up questions if you want, but I would not advice to do that by the reasons of compromised security I explain.
--SA

Sergey Alexandrovich Kryukov 28-Jul-11 2:40am

Very good points, my 5.
However, there is a way to use speech recognition I described; please see my suggestion.
I warned OP that security can be greatly compromised.
--SA

Wonde Tadesse 30-Jul-11 17:46pm

5+

Sergey Alexandrovich Kryukov · Accepted Answer · 2011-07-27T20:39:00

Totally wrong approach! I agree with Griff. Your problem is fairly simple.

Store text password in your database. Do it in hashed form using some cryptographic hash function, see http://en.wikipedia.org/wiki/Cryptographic_hash_function[^]. Use one of the functions from the SHA family, see http://en.wikipedia.org/wiki/SHA-2[^]. Use one of .NET SHA* hash functions, see http://msdn.microsoft.com/en-us/library/system.security.cryptography.hashalgorithm.aspx[^].

For login, use voice recognition. You need to use one of the two classes: System.Speech.RecognitionSpeechRecognizer or System.Speech.RecognitionSpeechRecognitionEngine, see http://msdn.microsoft.com/en-us/library/system.speech.recognition.aspx[^]. Using speech recognition is fairly simple; you will find some good code samples.

When the password is recognized, hash it and compare with hashed password.

Important note: this activity will give you very week security. First, anyone can hear the password. More importantly, recognition needs grammar. You can use so called DictationGrammar which is available and supplied with the engine, but it is huge and will give you a lot of false recognitions. You can supply smaller grammar, say, 100-200 candidate key words, but correct password should be included. At the moment of installation of grammar, all candidate words will be presented in its original forms, not in hashed form. If someone could possibly retrieve this grammar from where it's stored, the password could be obtained by trying all words in the grammar. You could improve it by generating unique grammar on the fly for each login, but the mere fact that the correct password is to be known to the system (which is not the case in hashed-only approach) would compromise security. In all cases, you should understand security very well to provide reasonable secure scenario.

—SA