Buenos Aires in Pictures

Hotel Carsson, Viamonte, Buneos Aires

Lobby Hotel Carsson

Lobby Hotel Carsson

Menu - Spanish makes it difficult for one to order.

Will Freshers + Seniors Combine enhance productivity in an Enterprise?

Work Performance is a Double Humped Curve

Introduction Grandchildren and the grandparents make a great pair except that there is this small hindrance that come in between them, in the form of parents. It is not difficult to guess why this is so. It is primarily to do with the "time at hand". While the "grand's" have a lot of time at hand, the parents in between are the greatly cursed for time. The overall problem, there is always a "parent-children"; "parent-grandparents" incompatibility and friction.

What is this to do with the work? Well, the "child-parent-grandparent" relationship and tussle follows almost like a shadow into the professional work environment.

Generic People Performance: A persons performance is a double humped curve. If we were to plot the performance of a person as a function of his age, it turns out that it is a double humped curve, atleast in the Indian context! Soon after college when he joins an enterprise, he is bubbling with energy and has all the time in the world and is focused on work; as he grows in the job, he is likely to get married, a dip in his performance (spend time with his better half), a little down the time lane, children into the family, further dip in the performance and this continues till the kids get to college, when the performance increases because there is not much work to do at home and the performance dip only with the onset of superannuation. See the figure below that depicts this!

Performance as a function of time of a person.

The Enterprise Problem: Consider a conventional work place of yesteryears (things have changed with young enterpruners appearing on the scene). A student who finishes his degree (for the sake of argument - Engineering) enters into an organization, call him a fresher. In all probability he is assigned to a project which is being day to day managed by a mid-level manager (Mid Manager), who in turn reports to a senior manager. For day to day interaction, it is between the fresher and the mid-level manager (the hump and the trough) and between the mid-level manager and the senior manager (a trough and the hump). Clearly, this is the parent (mid-level manager) sandwiched between the fresher and the senior manager, along the lines of the "child-parent-grandparent" relationship. The source of the problem is the time and or the lack of it.

While the fresher and the senior manager, albeit for different reasons, are flushed with time, the mid manage is slashed of time. And fundamentally there is a mismatch when we put the mid manage to interact with folks flushed with time in terms of amount of time that they have at hand. The best solution is to put the freshers with the seniors, atleast on the time front they are in sync. This might lead to a different kind of a situations and questions? Questions like what do we do with the mid managers? 

But then that is a different story :-)

God Lives in Argentina

Does God Live in Argentina?

An accepted paper in the IJCAI 2015 workshop meant one of us had to travel to Buenos Aires to present our paper. Because of the circumstances it was decided that I would travel to present our paper which spoke of a bio-inspired approach to correct the speech recognition output.

The first impression you get when you scan of BA on the web is that of an unsafe place with loads of information about where to use the ATM, worry about counterfeit notes, leeching taxis, variable currency conversion rates. Add to this the travel.

Mumbai to Buenos Aires is equivalent to almost 21 hours of -actual- plane travel. A short leg from Mumbai to Doha (3.5 hours), a long stretch from Doha to Buenos Aires (17.5 hours) with a short breather at Sao Paulo to stretch your legs. This is a journey that seems less enjoyable and more of a punishment if you traveled economy!

Reached BA around 830 pm and hunted for a place where I could convert my USD into Peso. I quickly figured out that English in not a language that the locals speak and I do not speak an iota of Spanish. My ability to communicate is tested, for the next 7 days.

I land up in an expedia.co.in booked hotel poorer by 450 Peso. Check-in Spanish-English and lo it is winter in July. Yes, BA is in the southern hemisphere. I looked up weather and I am made to believe that it will rain on my third and fourth day of stay in BA.

Being used to “predict rain” and no actual rain and vice-versa in Mumbai, my common sense tells me that I should keep an umbrella handy. I am ready.I look up the hourly weather prediction and find that the it will start raining from 6 am till about 1 pm and then there is sun shine. I tell myself, lets us. I wake up at around 630 and I find it raining. Wow. This is neat prediction. So I go about the day with my umbrella which means a walk of 1.5 km from my hotel Carsson on VanMonte Road to the conference venue – Sherton. And during the lunch break (around 110 pm) I sneak out and find that it is bright and sunny. Wow again.

And the correctness of the hourly prediction was to the hour throughout my stay in Buenos Aires. Making me believe God lives in Argentina. Na. God lives in Buenos Aires.

MH03 Registration? Repaid the Vehicle Loan? What Next?

Many, if not all, buy a car or a bike on loan from a bank or a lending institution. Typically, the paper work is done by the car dealer and there is usually no need for the buyer to visit the RTO (regional transport office) in person. Unless you want to!

However, one of the things that happens when you take a loan is that the vehicle is (hypothecation of the vehicle, namely endorsement of the hire-purchase agreement) in the name of the loan provider, obviously because he has lent you the money. 

You get to own the car once you have paid the loan. There is a certain joy of paying off your loan. But how do you get the ownership back? Simple, you need to "terminate of  hire purchase agreement".

My vehicle has a registration MH03. I recently paid off my loan and this is what I did. Hopefully this will help someone who wants to terminate the hire purchase agreement and has a vehicle with registration number starting MH03. I am not sure if the steps are different for different RTOs but more of less they should be following similar patterns if not exactly the same steps!

1. Once I had paid off my loan, I received two copies of Form 35 (Notice of Termination of an Agreement of Hire Purchase /Lease /Hypothecation) at my old address (we had changed our address after buying and before repaying the loan) duly signed by the lender which was valid for 90 days.
• Note: If you are not able to go to the RTO within this 90 days, you will have to get the Form 35 revalidated by the financial institute before you visit the RTO. A visit to the financial institute might be required. So best is to try and finish of the termination of the hire purchase agreement within 90 days!
• A visit with an invalid Form 35 to the RTO if not useful anyway!
2. Though there is information on the web, since there are two RTO's applicable for MH03 (see link) I emailed  RTO to find out where I should submit my form (Worli or Wadala) and what are their working hours. I received two helpful answers
1. "We do not work on 2nd 4th Saturdays. Office time for public attendance 0945 hrs to 1730 hrs."
2. "You can submit Form no 35 in Duplicate sign by financier and reg owner along with valid insurance, PUC certificate and registration certificate along with Fees at RTO Mumbai east office located at Sir Pochkhanwala Rd Worli. On any working Day  between 10 am to 2.30 pm preferably in first hr to avoid Queue."
3. Got the papers ready as per the details at link
1. Two copies of the Form 35 sent by the loaner (came with the seal and signature of the financial institute and I had to sign them too)
2. One copy of Self attested Photocopy of the valid PUC of the car
3. One copy of Self attested Photocopy of the Tax certificate (if you dig through the car papers you will find this receipt)
4. One copy of Self attested copy of the valid insurance papers
5. One copy of Self attested copy of my Registration Certificate (in my case it was a smart card)
6. Original RC.
4. I reached the RTO office in Worli at exactly 10 am. It was an effort trying to find the place because the office did not have a sign board! I had to ask folks standing around to check if I was at the right place.
• Thanks to the recent drive on agents in RTO, I spotted only the actual owner of the vehicles come for their work.
5. As you wait for the gates to open (yes the gate is closed and you can not enter the campus before 10 am - this looks alright when agents were allowed - by with actual owners coming for the work it looked unfriendly!) you need to sign on a piece of paper kept at the entrance to mark your token number (mine was 9 and I learnt later that there was not much crowd that day - I figured out that some people were there since 0830 am.) 
• BTW, you are not allowed to enter the premises until 10 am, you have to wait on the road - luckily there are plenty of trees and it is quite cool (my visit in May 2015).
6. Once inside
1. You stand in one queue and tell the officer what is the purpose of your visit. He will ask you to enter your details in a register and give you additional two forms and ask you to paste your photograph (this is where several of them get hit - because no where it is mentioned that you need to carry a passport photograph of yourself!) on one of them. And direct you to another queue for  paper verification.
2. The two additional forms are (a) details of the vehicle entered again by you - guess for the quick reference of the office that will process your form and (b) to ascertain who is submitting the Form 35.
3. You enter the next queue. The second officer diligently checks all the papers. Mainly looking for the validity of the Form 35, validity of the insurance papers, validity of the PUC and the tax certificate. He signs the paper and make a note of the charges to be paid on your papers + he nicely staples all the papers together. It was INR 100 + 50 for postage. He will direct you to go to Window 13, then Window 7 and then Window 8.
4. At the first window - the gentleman checks in his huge bound volumes (books) and  (I guess he has the list of vehicles that are under loan?) initials your papers and asks you to pay the fees in the next window.
5. You queue at the second window and you pay the designated fees to a lady; who  issues you a computer printed receipt and asks you to visit the third window.
6. You take you paper bunch with the fee receipt to the third window. He tears off a part of the receipt and attaches to the paper and gives the other part to you (this acts like an RC book until you get back a new RC book from the RTO) and he asks for your RC card which he puts in a plastic pouch and attaches to the forms.
7. At the third counter you are told that you will receive the RC (smart card) after about a month.
8. Done.

All of this took me about 45 minutes. I understand that I was lucky because there were not many people that day at the RTO.

Social Media as a learning tool

Probably one of the reason the discussion about social media as a learning tool is important is because of the popularity of social media. The popularity score is more in terms of active participation of the number of people of all age groups.

It would be wonderful to have this platform used as a media for learning.

Can Social Media act as a platform to assist learning?

1. Platform for Communication
2. Allows for active participation (in sense of user generated data; noisy data)
3. Generally all are equal (peer communication) but
4. some emerge as leaders (may be based on their expertise, participation, creativity ....)
5. SIG formed (Leaders moderate the group)
6. leads to some focused discussion (signs of learning - though not formal)
7. since there is less noise "useful learning can happen"

So structurally everything is in place to enable learning if no distinction is made between formal and informal learning!

With some constraints on the social media - allowing a teacher have some special privilege to moderate; like allowing or restricting "what to do when"?  The social media as a platform and an ICT learning tool can be stitched together.

However, any constraint (hierarchy) reduce the "social media as a learning tool" to a mere classroom (CR) kind of environment. At best one could call this social media learning tool as CR++.

Now the question is should CR++ be in addition to a class room or in its own right an independent platform for learning? Independent of the answer,  CR++ could have the following advantages:

• Multiple teacher driven courses (different versions of the same topic)
• Multimedia driven courses means may be faster learning and stronger understanding
• students learn anytime anywhere
• if you missed CR, use CR++ or if you did not understand CR use CR++ to refresh general problems
• Requires infrastructure (communication channel, devices) for all participants
• assumption that electricity and bandwidth is reliable)

If you dilute the definition of social media a bit and look at it as a platform to make things immersive, faster information delivery then one can look at Khan Academy type of learning. However for this to work .. some kind of formalism needs to be planned by educators especially in terms of content generation which address

• Culturally - (date-month vs month-date)
• Geographically (left hand drive vs right hand drive)
• Language (several non-English!)
• rural learners vs urban learners
• Higher, Middle, Primary, Play school
• learners with disablity

Some  platforms that can be gainfully used for learning.

• FB/Orkut
• Twitter
• Voice Twitter
• YouTube Channel/Podcast/live cast
• LinkerdIn
• "X" Groups (Google, Yahoo)
• Blogs (WordPress, Blogger)
• Wiki
• Skype

Endnote: SM is more immersive (something like a Radio versus TV). Probably too early to infer how social media can be used as a learning tool. Like the "missed call" came out of necessity (hole on the wall experiment).. may be something will come out out of sheer necessity.

May be travelling becomes too expensive (with the traffic jams!) or time consuming meaning CR++ is the only option!

Traffic in General

"There is a traffic jam on a particular road and hence take an alternative route, now enjoy the music" is the general message on the FM radio in the car.

It set me thinking, I hear this almost everyday when I am driving in Mumbai, India and yet it is just treated like any other news ... or may be not. It has sort of become "no news" because it does not contain anything that I already do not know. Probably the day there is no traffic jam will sort of become the news!

It is a pity that we (all) know it yet we are either unable to do anything (which I hope is not true) or probably think what can I do about it (reasonable to think this is true)?. 

Why is it that we are prepared to spend time in the traffic day in and day out and still not worry about it? Technology advancing, IQ of people increasing, everything seems to be progressing (economy?) and so is traffic!

There are several though processes. Several of them blaming the government. Popular among them 

1. Decrease the number of vehicle licences, 
2. Provide better public transport, 
3. Hike the price of petrol, ...   
4. Allow only even number vehicles to run on even days etc. 

but we have seen none of this has taken shape. Obviously government does not see this because the people who can do something (call them VIP's!) actually zoom through on the road with the traffic constables (posted almost every 20 m!) making sure no other vehicle is on the road so that the Minister thinks the road is traffic less!

But the traffic remain. Probably it is up to us to try and see how we can make Mumbai roads "traffic jam free". How? There is no one magical solution I am sure.

Train Journey in the 70's

My father, a central government of India employee, made sure we made our religious LTA (leave travel allowance) when he was in service. Thought there was no IRCTC, my father almost always made sure we had all the tickets reserved in advance, which meant a trip to the railway station and long wait in the queue and with a finite probability of not getting your tickets after hours of standing in the queue. 

Once the tickets in hand, these journeys those days used to be great, especially because my father used to make it a point to write to all his friends about our train travel, the date and the likely time it would hit the station close to them. 

Invariably, we had someone or the other come and meet us along the way at the stations that our train used to stop, so this once in 4 years LTA meant we did meet our friends and relatives albeit for a couple of minutes on the platform. 

As kids the most interesting aspect, more than just getting to meet them, was the breakfast, the lunch or the dinner that his friends used to get; it was not so much not to eat outside -junk- food, but it was more to do with this aunties food or that friends breakfast. The well cooked and packed food was a show of friendship and in a way was a substitute for the warmth and affection that they would have bestowed on us as their guest were we to park at their place for a couple of days.

On one such LTA trip in mid 1970's my father had planned a trip to Jammu and Kashmir. As usual he had written to all friends who lived in towns and cities along Bangalore to Jammu via Delhi. The ongoing was as usual great, the train was in time (couple of hours of delay was usual!) and we met all our friends; to us kids, it was the food. On our way back from Jammu and Kashmir, we got to Delhi from Jammu and figured out that the train that was to take us back to Bangalore was a non-starter and as a substitute a couple of extra compartments were clubbed with another train; while this was comforting; however in this process we found our compartment (the extra bogie) to be almost invariably not on the platform or so far from the shops on the platform when ever the train halted at any station.

At Delhi, because of the uncertainty, my parents did not think of the food we were to eat during the journey and as night fell they panicked because we the kids were hungry. There was no food in sight and there was no pantry on the train. As night fell, my mother tried to put us to sleep on an empty stomach. It was, as often recollected by my mother, when we the kids were asleep, the train stopped yet again at some station. But luckily this time, she saw a group of people on the other side of the platform (they had booked a compartment for themselves and it was halted to be tagged onto another train the next day) and heard them speak out tongue; she walked to them and saw them finishing of their dinner on the platform. Very shortly she figured out that they were pilgrims and quickly the pilgrims got to know that there were these two hungry kids who were asleep.

What I as a kid remember, is the instance, where my mother is asking me and my sister to wake up with hot rice mixed with daal, mango pickle and ghee in front of us. I think that was one of the most fulfilling dinner that we had as kids.

To this date, we always recollect this incident, the kindness of sharing food by complete strangers in a distant platform.

Ascertaining Person Behind the Voice Seamlessly

Background

When I transact online today, there is always a need for me to authenticate myself. Be it to see my mail, be it to check my account balance or order a pair of shoes for my son. If the mortar days, you visited an enterprise between a fixed working hours, at a specified location and interacted with a set of people representing the enterprise in person. The only form of authentication would be your signature or your thumb impression; however informally the authentication would be at a deeper level. For one the person representing the enterprise actually got to look at your face, your body language and spot if anything was amiss. The signature or the thumb impression on a piece of paper was more of a formal book keeping exercise, just in case it was required in future to verify a transaction.

As the enterprises broke the barrier so as to allow their customers to do anything anytime, it was welcomed with wide and open arms by everyone. It was convenient; one did not have to travel; did not have to plan to visit during a certain time to do a transaction. But this came with one major problem; how is the person on the other end to be authenticated. It started with choose what ever you want as your password, if one transacted on the web, but very soon went on to put several restriction on what could be your password – can not be a dictionary word, use atleast one number, use capital letter plus a number, atleast 8 digits. All this to make sure no one else impersonated and tried to transact as you. When people questioned, if I was transacting with the enterprise that I want to transact with actually gave raise to two factor authentication and so on. Strong, unambiguous, robust authentication was required because there was no physical presence that was one of the main aspects of a mortar transaction.

As the digital space widened, people presence in the digital space increased. People has to remember several passwords. They had two options, keep the same password everywhere and ride on the risk of lost password would unlock all your digital space or have a different password and stand the risk of forgetting passwords. What if one used something very specific to oneself as the password? Then there would be no situation of a forgotten password because one did not need to remember it anyway. Step in biometrics. Biometrics became popular, when in doubt one popped up a question related to you, like your date of birth or you place of birth or your mothers maiden name – use of biometric was and is still used effectively to retrieve our forgotten password.

I like to Talk

Though there are several digital channels, like, telephone, email, web FAQ and SMS that allow a customer to interact with an enterprise. Some studies indicate that a majority of the customers choose the telephone channel to interact for any customer care support. Especially customers like to speak to the human-agent (see Figure below).

Figure 1. Customers Like to Speak to the Agent!

Further it has been observed that there is a higher customer satisfaction for a phone based interaction (see Figure 2) compared to other existing interaction channels. While there could be several reasons for this, it is not hard to guess the possible reasons for a higher customer satisfaction in a telephone interaction channel

• Telephone channel is more personal,
• Telephone channel enables live and real-time interaction with the agent at the other end.
• Speech is the most natural mode of communication to sort problems
• It is a synchronous transaction
• Closest to a face to face interaction.

Figure 2. Customer Satisfaction

When you transact on the telephone two things happen, you are doing your transactions live and in real time and at the same time you are not bodily present with the person serving you. So the person representing the enterprise has to ascertain that you are indeed the person you are claiming to be without actually getting to see you – authentication became very crucial. The telephone channel of communication meant that your voice is the only cue he can sense as he interacts with you.

One of the easiest thing to do, which was done by several enterprises and which continues to be done was to ask for your data of birth, which you either spoke or entered through the DTMF of the telephone keypad. The person on the other side heard your biometric information and compared it with your data that was available with the enterprise to ascertain that you are indeed who you claim to be. While this was good, it was not sufficiently robust, what if people somehow get to know your date of birth and used it to impersonating you to gain access to your information. Additionally, enterprises started looking at technology that allowed their customers to help themselves rather than have them speak to a human agent. The presence of technology and the cost implications probably were the two main reasons to go the self help way.

Help Yourself

Self help allowed you to interact with a machine, except that the machine drove the interaction. Popularly called the IVR based self help systems. “Please choose an option. Press “1” for .. Press “2” for ...” and it responded to key presses; when there was a need to authenticate you it asked you to use the telephone keypad to enter your date of birth in a certain DDMMYY format.

However this interaction was unreal without a human touch, guess we missed our natural instinct to speak to resolve a problem! Eventually you found that there was a Press “9” or a Press “*” to talk to our customer representative.

What started off from the enterprise perspective to allow you to help yourself and there by lower the cost of having an actual human talk to you failed to take off; when people had to use the self-service because of long wait in the queue to speak to the agent they did so but grudgingly and the customer satisfaction plummeted.

Although self help channel still existed, it was the conversation with the call center agent that was more popular. Enterprises jacked up their infrastructure, tried to find ways to reduce the transactional time so that more customers could be serviced in the same amount of time etc. Could Speech recognition technology help?

Talk and Help Yourself

Speech recognition technology made it possible for you to speak to do your transaction. The idea was that the machine which was interacting with you still took hold of the interaction but it allowed you to speak rather than ask you to press the DTMF key on the keyboard. So instead of “Press 1 for …. Press 2 for ...” you now were asked “Speak /cheque/ to order a fresh cheque book. Speak /account balance/ to know the balance in your account”. This was far better than the Press “1” for …. kind of self help.

One of the very important prerequisite in this kind of a voice based self help was the need to authenticate the speaker based on his or her voice. Enter Speech biometrics.

What is in What I Speak

There are essentially three components that are embedded when we speak. When I speak /good morning you are reading my article/ there are three things that can be sensed by the listener. The firs is the content of the speech, namely the text that I spoke (“Good Morning. You are reading my article”).

The first is the Linguistic information, which is primarily what is said. It could be the language in which it was spoken (English in this case) and gives an idea of what was said (the equivalent of speaking a written text).

In addition, it also carries some non-linguistic information like who spoke (me in this case!), how I spoke (hopefully pleasantly!), what is the emotional state of mine when I spoke it, gender of the speaker, the accent, the articulation clarity. Some of these are purposely added by the speaker and hence are controllable while some are inherent to the speaker and can not be purposely be controlled by the speaker.

Speech biometric exploits to a large extent all the non-linguistic information in spoken speech to ascertain the identity of the speaker. Essentially, there are two broad ways in which speech biometric is used, one is in the form of speaker verification, while the other is speaker identification.

Speaker verification is the process where the speaker claims he is X and the system uses an a priori model of the claimed person to verify the genuineness of the claim. The system measures the closeness of the speakers speech to the model of speech associated with that speaker and if the speech is close (based on a preset threshold) it validates the identity or denies validation (forgery). In a speaker verification system, the decision is binary in the sense that once the speech of a speaker is compared to the speaker model the decision is one of YES or NO. This decision is usually based on a threshold value. The choice of threshold is determined by the type of application. Primarily there are two contradicting measures which have to be kept low for the success of a speaker verification system, namely false acceptance ratio (FAR) and false rejection ratio (FRR). While the FAR refers to the percentage of forged attempts that have been validated by the system, FRR refers to the performance of the system as not being correctly able to validate genuine users. The choice of threshold has opposite effect on FAR and FRR. A choice of threshold might minimize FAR or FRR and not both simultaneously. In brief, a threshold which has a good effect on FAR (lowering) has the reverse effect on FRR (increasing) and vice-versa. A very secure system would choose a threshold such that it would have zero tolerance to FAR at the cost of an increased FRR (rejecting genuine speakers).

A quick browse of Wikipedia tells that, speaker verification systems come in various forms based on the amount of security that is expected from the verification system. Broadly we can categorize the speaker verification systems into four categories, namely, (a) Fixed Phrase system where a pre-determined phrase is used for verification, (b) Fixed Vocabulary verification system is more flexible and practical; training and testing materials for a speaker are generated based on words of a fixed vocabulary. The user speaks the words or phrases from this vocabulary both at the time of enrollment (or training) as well as when testing. (c) Flexible Vocabulary system use a general set of sub-word phone models which are created during speaker model training (d) Text-Independent system, where the user is not constrained to say fixed or prompted phrases. He has the freedom to speak anything and the system works by extracting the speaker characteristics from the spoken utterance. Clearly, both complexity and security increases as we go from fixed phrase to text-independent.

On the other hand, speaker identification is the process of identifying the person behind the voice without the person specifying who (s)he is.. For example, in banking scenario, if the banks customer base was N unique people, then when a user calls, a speaker identification platform would have to first determine if the speaker is one among the N people that it knows and then if the person speaking is one of the N then it has to identify which of the N is the speaker. This is essentially a problem of choosing one among the N+1 possible identities. N for the people and the 1 represent not among the N people!

Clearly, speaker identification is more challenging technologically than the speaker verification system; as a result most forward looking enterprises are looking at speaker verification system only.

Using Speech Biometrics

Services as an industry has had a significant growth in the recent times. With this growth there has been a need to service a large number of clients which is on one hand very important but on other hand a significant cost to the service industry. Many service industries have tried mechanism so as not to dent the customer satisfaction while trying to minimize the cost of providing a channel for its customers to interact with them. Automated self help based solutions are the front runners and the telephone (IVR) channel is the most sought medium of interaction.

Enterprises can provide personalized and accurate information if they know whom they are servicing and one of the non-intrusive method to identify the person on the telephone channel is through spoken speech. In this context speech biometrics becomes very important.

Speech signal as a biometric is gaining significance in recent times. Speech biometric is probably the most non-intrusive and natural biometric to verify the identity of a person. The use of speech biometric even in the speaker verification mode is plagued by the well known fact that it is not cent percent right cent percent of the time (specifically when only speech is used as the cue). So there has been certain reluctance to use it in practice. However there has been several commercial speech biometric solutions available in the market claiming usable accuracies of verification, in almost all the cases they use more than the speech cue to verify the identity of the speaker. For example, one of the important cue would be the need for the user to authenticate himself only from his registered mobile phone number.

Irrespective of the performance accuracies of the speaker verification system based on the speech cue, most implementation today speak of replacing the authentication process by speech biometric system. For example, today when I call a service provider, after the initial welcome message (“Welcome to …, How may I help you today?”) you are posed by a simple looking question by the human agent (“Am I speaking to …... can I know your date of birth please”) which is what is used to verify my identity.

In a self help kind of a scenario, the authentication is much more direct. Sensing my registered mobile number, the system knows who I am supposed to be; the system blurts out “Please speak your password”. I speak my password, most probably a password that I resisted with my service provider. If I am verified I am allowed to do the rest of the transaction else I am asked to leave or if the service provider is courteous I am put in a queue to be helped by a human agent.

While there is nothing wrong with the use of speech biometric as a replacement for the actual human authentication process. Clearly, there is a customer dissatisfaction, imagine a machine refusing to identify me and then putting me in a queue to speak to an agent. If this happens several time (two to be exact!) I will not opt for anything other than speak to the agent!

The question is can I use speech biometric, with its known problems of not being accurate all the time such that it is transparent to the user so that (s)he does not get a feel that (s)he is being verified at all.

Invisible Speech Biometrics

There are two things that happen when one used speech biometrics in the form that it is envisioned to be used. It not only disturbs the flow of interaction, the speech biometric authentication step stands itself as a process (eating away precious time to just authenticate the speaker); additionally (a) it is a point authentication, meaning the authentication happens at one given point of time (what if the phone changes hands after the initial authentication?) and (b) there is a customer dissatisfaction if the speech biometric system fails to recognize a genuine user. Can we do better?

The author in his ignorance, anticipates the use of speech biometric to handle all the drawbacks of a regular speech biometric system (see Figure 3). The idea is that of continuous authentication from the beginning till the end of the call there by eliminating an explicit “authentication process”. See the marked “Time Jump” in the Figure 3; which means there is no need to explicitly have an authentication phase in the call interaction. The continuous, in the background speaker verification means that the process of authentication is invisible to the customer and hence the chance of dissatisfaction does not arise. The third issue of phone changing hands in midst of the transaction could still be identified.

Figure 3. Continuous Authentication

Additionally this process of invisible, continuous speaker verification means that the enterprise can still give me access to certain things even if it verifies me with with low confidence – for example if my transaction is just to verify the change of address that I had sought … then the enterprise can give me the information even if it does not verify me with high confidence … but when I want to transfer some money it might want to be able to verify my identify with higher confidence. This gives the enterprise the scope to only verify me when it is required but also decide what degree of confidence is required for what kind of information.

Conclusion

Speech biometric can really make inroads into actual use if it is used with its known limitation and the understanding of the positive or negative impact that it can have on the user. The speech and natural language team in TCS Innovation Labs – Mumbai is exploring on how best to provide a speech biometric into an already existing setup where one could avoid the point authentication process by continuously verifying the speaker. We believe a continuous rather than a point authentication is going to be the way forward because it addresses the challenging that come with the current though process in implementing speech biometric in a call center or otherwise.

(TL)ea(cr)(hn)ing

(TL)ea(cr)(hn)ing

The title (TL)ea(cr)(hn)ing is by intention. Without rubbing much of our brain it spells out Teaching and Learning in a single line.

Teaching and Learning are two sides of the same coin. Someone not so famous said, can I learn without teaching and can I teach without learning. Guess you learn even more as you teach and you teach only when you learn.

Learn anytime, anywhere is the current trend with all those online courses being made available. Initially it was just the digitization of mortar courses and not it is using multimedia to stream the concept. People say that we have moved from a stance where it was bookish and limited to a few to a more open leaning system which is available to almost anyone and most of the times “free”.

I still believe in a Guru shishya kind of a setup, where a student served the guru and in return may be he was imparted knowledge. The amount of effort that a student put in, in some sense displayed the thirst for knowledge and when the knowledge was imparted it was absorbed in spirit and body. I wonder if the leaning that the students derived would be as intense or as absorbing if the knowledge was imparted the day the student joined the guru. Today the stance is you have information (not knowledge) on the platter, it is for you to absorb the knowledge, you absorb the quantum which is determined by your focus and need.

Information is available and not knowledge is a fact. The upside of this is you have so many sources of information – but to try and find the information that is backed by an expert is often not considered. Just imagine a learning material put up by a person on effective cooking who has never cooked! The other aspect of this is that when there are several choices you need to appropriately select and quench your thirst for knowledge. In the good old days it was the teacher who did that; she acted as the funnel; disseminating information appropriately. If you were a bright student probably the outlet of the funnel was larger else it was smaller. Today in the online learning this important aspect of funnel is missing.

The anytime anywhere extension of the knowledge learning has a problem in a very different dimension. Discipline. 0830 morning lecture means I need to be in the classroom at 0829 and I plan and be there. But the anytime anywhere makes you less disciplined. If not now I can learn later, may be after a delay of 30 minutes, 1 month, 1 year of never! But the aspect of discipline is definitely at risk. While there is a notion that online and non-mortar courses are free a fiercely detailed person would immediately suggest that there is a cost involved in terms of data download and power used to run your device etc. My personal opinion as far as learning is that When you put effort - time and money you usually learn better.

Appropriateness of information. There is too much of material that might not be applicable globally. While I can not think of something in the college learning, I recall buying a book for my toddler son. Guess it was on learning manners or something equivalent, and when I was on the page that spoke about manners on the road, it said “always walk on the right side of the road”. Well it was a right statement expect that it was correct for the countries that drive on the right side of the road and not on the left like what we do in India. I immediately discarded the book. But the learning from this is “use local material” to (TL)ea(cr)(hn)ing.

My own vote is for classroom teaching (I do miss the chalk dust; the screeching noise of the chalk on the blackboard). The aspect of group learning is much much different than learning in silos. The questions being answered almost immediately versus the need to search for answers when you take an online course are some of the reason.

Learning for marks or certification is probably “just sufficient” learning so that you get the certificate or pass the exam. But the real learning will only happen when there is no pressure of marks or a need to get a certificate.

Happy Teachers Day.

What is in a name? Number?

When we wanted to name our son, we did a Cntrl C Cntrl V (Cut-Paste) from our names and made up a name and thought that we had come out with a brilliant idea. The Cut-Paste worked well for us to get to a name that was (a) very simple, (b) easy to spell, with just five alphabets. No surname just the name. By the way we did get his passport also with just his name!

SUnilnaMITa

It is another thing that it was not very visible that our names was hidden in our son's name so we had to tell people we knew specifically how we had named our son. While some said we were lazy, some others took a cue and tried to name their new born the same way. So there were some compliments and there were some jeer. When we told our son about the way he was named, he insisted that he was not sure he liked his name! 

However as far as I was concerned. I was under this illusion that we had done a great job by thinking of a CntrlC-CntrlV name.

This was some twelve years ago.

My tweleve year old son came to me the other day and asked me, 

“Can I please have a personal email account?” 

He insisted that it be a gmail account, not very sure why though.

After a while I figured out that it was not possible to get an email account which just had his “name”. Probably it was taken. While I tried several options and thought that “name.eman” which incidentally was available; was a fantastic idea. But as should be, he was not keen on a 2 times cut-paste name. Finally he chose “name.school_name” as his email. You see he loves his school!

Set me thinking will there be a planned name in which parents will choose their kids name? I my sincere guess is yes. What will the choise depend on? May be the availability of an email address?

Couple of years ago, the name of a child, especially in India, was constrained by several beliefs. A name definitely had to have the 

• family surname, 
• the family deity name could not be left out and displease the gods; 
• well known ancestor's (grandparents) name was deemed good to have, 
• village name so that people knew where you came from, 
• profession and 
• what not

after making space for all these, one included the -real- name by which the kid would be called. 

This method of appeasing all meant the names got pretty long and in a way unique. The upside of this in the age of internet was that it allowed one to obtain a login of your choice on any e-commerce site that you wanted to including an email account! 

However over a period of time the general trend was to shorten ones name. People questioned, why should I carry my profession in my name? Why should I have a deity’s name in a name etc. This ended up making the names shorter which in turn makes it less unique.This was alright before the internet, email, e-commerce were popular. But now with turn in increased e-commerce activity, one needs a unique name to transact.  Else you will need to choose a name with numerals which does not look all that representative of a person!

This set me thinking, will parents now work on a different set of constraints to name their children? Will naming of children be a -new parents- dilemma?  Short or long name?

• Will it be based on some uniqueness metric? 
• Will it be based on ones ability to get a decent name space in the digital space (so that one can get a gmail, a facebook account or a twitter handle)[*]? 
• Will names be auctioned? The ones that can get you a nice email address would cost more?

[*] assuming that people want to use their real names even in their virtual world. 

Or will all of us be reduced to a set of numbers? Like your employee number, your bank account number, your UID/SSN? May be a  universal identification number. The first two digits to signify the nation, the next two to .... and the last x digits represent you.

Atleast kids will not crib that I do not like my name! Or will they say, Wow. My number is a prime? 

How do I Choose my Undergraduate (Engineering) Project?

Let us face it; all of us as undergraduates have faced the dilemma of what project I should choose as my final year Engineering project.

Recently I was asked this question by a respected faculty of an Engineering College in Mumbai. "Can you please speak to our final year students on how they should go about choosing their final year projects?"

I was not too sure if I knew myself what was a way, leave alone the best way, to choose a project. As I gathered my thoughts, the first thing I figured out is that it was not a synchronous process which could be put as a sequence of slides.

Here are my (not so comprehensive) thoughts in a picture :-)

 

 Am sure there is a bias towards the work that I do, however this picture (created as a MindMap using freeplay on Ubuntu and exported as jpg) should enable a final year engineering student ask the right questions to converge onto a project (s)he would love to do.

Good Luck.

Online Trading - Phone is mightier than Web!

"Can you invest in L&T Emerging Business Fund? It is a NFO and the last date is tomorrow" asked me.

I thought it should be a zippy thing to login to my trading portal and click a couple of buttons and it should happen. And I had not bothered to acknowledge the existence of telephone trading. I found that I was wrong; we are not yet done away with the mighty telephone. Read on ...

I login to my trading account portal. After a little navigation I find that there is no option for an NFO under the L&T AMC. I do the next best thing .. start a chat. The chat is slow but finally I conclude that the only way this order can be placed is through telephone. I ask for a number and he chat "append your local area code and dial 3*5*3*6*". Is there a 1800 number I chat back? "No sir this is the only number".

I dial 3*5*3*6*, choose "1" for English option, the first thing it asks me type in a TPIN which I do not have (thought I would only use the Web!). Luckily I trace back on the IVR and get to an agent.

Agent, "Do you have a TPIN. No? I will help you generate one. Tell me your DoB".
Me,"19xx"

I am now in a three cornered conference. Me the agent and the Computer (IVR).

The IVR asks, "Enter your DoB"
The agent enters this for me! (saves my time!)
IVR,"Choose a 4 digit TPIN"
Agent,"I will go offline you can choose a TPIN"
Me,"DTMF, DTMF, DTMF, DTMF"
Agent, "You will have to type in again" (not sure if the agent was listening to my DTMF!)
Me, "DTMF, DTMF, DTMF, DTMF"
Agent, "Now you can tele transact"
Me, "Thanks".

Now the IVR puts me in touch with another agent. Who helps me carry out the transaction. As I wonder how they can confirm that I had actually asked for this transaction ..

Agent,"Please confirm. You requested for blah blah ... Do you confirm"
Me, "I confirm the transaction"
Agent, "Thank you. You transaction number is *****"

End.

This could have been done much better on the Web. Unfortunately, this particular NFO was not listed on the portal of this trading company!
 

Datasets

Datasets

Def: A dataset is a collection of data or a dataset consists of all of the information that needs analysis, generally the data is gathered during a survey.
 
You have great idea. You want to conduct some experiments. You collect your own data, test your idea and report results. Most often the data collected locally is small (to suit your experimentation).

You submit your results and then there is a comment from a reviewer saying that  (a) your experiments have not been tested on a large public datasets and (b) you need to compare your results with existing literature.

You can address (b) by implementing the existing algorithms and then running these algorithms on your small dataset. Obviously more work (need to implement an already known algorithm!) without much returns.

Use of publicly available datasets is a a good idea. It addresses both (a) and (b)!

1. You get to test you idea and experiment on a large public dataset
2. You can compare your results with existing literature (assuming that they have also reported results on the same dataset)

Here are a list of datasets that can be of use

• Public Data Sets Amazon 
• UCI Machine Learning Repository: Data Sets
• Small TSP Dataset
• Climate and other datasets
• List of dataset for Data Mining
• Statistics Dataset (includes breakfast cereal dataset!)
• Stanford Large Network dataset
• Large list of dataset
• ChestXray dataset from NIHCC

All said and done, as a user of these datasets one should also strive to create datasets and put them up for public use!