Amazon Echo Plus in India – first impressions


Got my first smart speaker, the Amazon Echo Plus yesterday and it seems to be nicely Indianised. It seems to be using the Raveena voice based on Indian English or a variant thereof from Amazon Polly (AWS’s text to speech service). The Alexa app itself is also pleasantly adapted for India with the appropriate command suggestions and services available.

Some of the India specific commands to try

I had pre-ordered the device and it was delivered within a day of the dispatch. The setup process was quite smooth and once done, I promptly went about installing a bunch of skills ranging from the utilitarian to the time pass ones.
As a music player, the sound quality is decent but nothing spectacular as many reviewers have noted. As for the music catalog, it seems to be using Saavn exclusively though it does Amazon music as one of the options in the app. Saavn itself has a decent catalog and Alexa has again been Indianised sufficiently to understand some Hindi song names. I  tried “Play the song ek main aur ek turn” and it actually started playing the song from Saavn though it did pronounce “main” the English way.

Alexa understands Hindi!

The flash briefing skill is quite handy to get a quick bulletin of your areas of interest once you have set it up with your desired sources. It can also give you cricket score updates without any skill installation as I tried out during today’s India – New Zealand match.


I installed the Uber and Ola skills to check the overall utility factor. While Alexa seems to be able to book an Uber including picking up your location, the payment mode defaults to cash which is a dealbreaker for me. Then there’s the Zomato skill I installed and tested. It seems to know your last 3-4 orders and you can reorder as well but didn’t go beyond browsing for the moment.

The alarms and timers work pretty well too and I conveniently set a sleep timer to stop playing the music while going to bed.

Then of course there’s the whole reason why I got the plus instead of the regular model which is the smart home hub built into the device. A solo Philips hue bulb is what I ordered next and setup today. It was again a pretty simple process with the Echo detecting the bulb in a few seconds. Controlling the bulb by voice is also quite easy right from switching it on and off to changing the colour and brightness. The app however has just the on/off switch and brightness control at the moment and as many reviewers have noted, the functionality of the smart devices using just the Echo Plus is considerably limited when compared to using the devices with their respective hubs. This is definitely one area of improvement and given the kind of coverage you see for smart homes on the Amazon Alexa pages, it should improve sooner rather than later.


Apart from all this you can also use the Echo as a Bluetooth speaker and pairing it with my iPhone was quite simple. That said, all the sounds from the phone start getting carried over to the speaker and this interrupts any song or speech playing on it directly. Due to this, I ended up keeping the phone disconnected unless I wanted to play something from my phone.

One thing I couldn’t find is the voice profiles option that lets Alexa identify the person speaking and customising the responses accordingly. Possibly a feature not yet rolled out to the Indian market as it seems to depend on the Amazon app that didn’t seem to have this option in my case.

A smart speaker is a family device but my wife is not very enthused by the idea while my 4 year old daughter would like to play with Alexa but hasn’t yet gotten out of her initial shyness phase to begin talking freely to her. It didn’t help that Alexa couldn’t answer many of her queries and also the fact that she was trying to get Alexa to identify the colours of the crayons she was holding in front of the device – a perfect case for Google Lens and Assistant.

This is of course just what I’ve been able to check out in the first 24 hours with the Echo Plus and I’m sure there’s lots more already available and also coming in the near future.

Advertisements

What a commuter wants…

… is a reliable means of transport to get them from point A to point B. Of course, it helps if the commute is pocket friendly, comfortable and fast. While taxis & autos in Mumbai adhere to the fare meters (unlike certain other metros), refusal is a universal problem shared by all commuters irrespective of the availability. The first generation of private cab services like Meru, TabCab, EasyCab did try to sort out this problem to an extent, but never managed to have enough cabs available or offer fares competitive with kaali-peelis or autos (AC notwithstanding).

Ola also jumped into this space following in Meru’s footsteps before significantly restructuring their pricing model upon Uber’s entry. These 2 taxi service aggregators reached near kaali peeli fare levels and offered a much more reliable (read disincentivised refusals) and more readily available service. In fact, Ola even tried to get kaali peelis on their app, but the effort seems to have fizzled out after a promising start.

The rest of the script is also playing out just like in the rest of the world and even a city like Mumbai, the so called commercial capital of India, has witnessed 2 taxi strikes within a couple of months. While the first strike was accompanied by Mumbai commuters discovering the basic Economics concept of supply and demand thanks to the Uber surge pricing, the second one has shown how disabling surge pricing makes life difficult and reduces the availability of cabs. Either way, the commuter has gotten the wrong end of the stick.

I just hope that we find a better solution than the other countries to this whole standoff between the incumbents and upstarts. Too bad the kaali peelis and autos don’t think of adopting a no refusal policy – something that’d get Ola and Uber in real trouble.

Reality catches up with Uber – Mumbai taxi fares re-revisited

Uber Mumbai has just announced a big hike on the Black and SUV services, pretty much bringing them on par with the Ola Prime SUV service. So here’s the latest fare chart (older versions here – v1, v2):

Approx. taxi fares in Mumbai as on 13 July 2015
Approx. taxi fares in Mumbai as on 13 July 2015

Note on the calculation methodology:

  • Travel time calculated assuming 3 min per km (Uber, Ola, TFS)
  • Waiting time taken as 1/2 min per km (kaali peeli & Meru\TabCab)

Revisiting the taxi fares in Mumbai

Ola announced a series of price cuts to their Mini and Sedan services to better compete with Uber and also added the Taxi for Sure hatchbacks to their app in the last few days. This calls for an update to the fare chart that I had made for the various taxi services in Mumbai ranging from the traditional kaali peeli and Meru\Tab cab to the new entrants like Ola and Uber. So here it is:

Taxi fares in Mumbai
Taxi fares in Mumbai

The equation hasn’t changed drastically, but the Ola Mini service is now pretty much comparable to UberX, while UberGo remains unchallenged. Ola Sedan also becomes significantly cheaper than the Merus and Tab Cabs while the newly added Taxi for Sure service (for the Ola app) slots in between these two. TFS seems ripe for a round of price revisions as the  cars are effectively equivalent of the Minis, i.e., hatchbacks.

The recommendations are quite simple:

  • For short distances (<10 km), kaali peelis are the most economical
  • Beyond 10 km, UberGo reigns supreme. In fact, unless you are doing very short distances (sub 5 km), they are the best option. They’re definitely not sustainable for Uber and that possibly explains their relatively limited availability. However, for taxi commuters like me they’re the perfect kaali peeli replacement.
  • Since you are unlikely to get an UberGo, your next best bet is to settle for an UberX or an Ola Mini. For that matter you could go with any of the other options barring the SUVs or Uber Black for distances around 10-15 km without too much fare difference.
  • For distances longer than 15 km, the newer lot comprising of UberX and Ola Mini & Sedan pull away from the Rs 20/km crowd of Meru, Mega, Tab Cab etc.

Either way, this is a good time for the commuter though the rates are unlikely to be sustainable in the long run. So, enjoy for the time being and hope that the day of pleading with taxi drivers and autos never returns.

How Uber’s shaken up the pricing structure in India

I’ve been using Uber quite frequently over the last couple of months and today’s Mumbai taxi strike to protest such services ironically forced me to opt for Uber at a 1.8x surge price. While I’ve had my share of ups & downs with Uber, the flexible pricing model has been one aspect that I’ve been impressed with compared to the competition like Ola.

Uber managed to create quite a buzz offering single digit per km rates which was almost half the rate others were offering at that time, but the pricing model which included a per minute charge on the trip ensured that the overall fare was not unsustainably low. This has also allowed them to go after the local taxi & auto services in the different cities and they also end up being cheaper for medium to long distances.

The Uber pricing in India is typically a low per km rate coupled with another per trip minute rate on top of a fixed base fare, with the overall fare subject to a minimum amount and of course the surge factor. Putting it simply:

Fare = Surge factor x (Distance x Rate per km + Trip time in minutes x Rate per minute)

Ola which had started off in India with a conventional pricing model of rate per km and a waiting time rate has pretty much overhauled their pricing to mimic the Uber model. They have in fact abandoned their initial method of applying a fixed peak time price during 2 slots on weekdays in favour of a surge factor. The other taxi services like Meru, Tab Cab, Easy Cab etc. have thus far stuck to the traditional model, though they’re trying to stay relevant through special offers.

I also did a simplistic analysis of how the different services compare in terms of the trip fare in a city like Mumbai (Google Sheet here). I’ve assumed a trip time of 3 minutes per km and waiting time of 1 minute for every 4 km, so the results are going to be quite different in heavy traffic.

Approx fare comparison
Approx fare comparison (corrected)

For short distances, the local kaali peelis are of course the cheapest, but for distances above 10 km, UberGO ends up being a better deal. The next cheapest is the Ola mini which starts getting pretty competitive with kaali peelis after the 20 km mark. This is of course disregarding the non-AC nature of the kaali peelis. [Update] Ola Mini and UberX are pretty competitive till the 10 km range, but separate pretty quickly after that as the near 30% higher charge per km for Ola starts making a mark.

The older generation of Meru, Tab Cab etc manage to remain competitive with the newer lot, matching the next best Ola Sedan UberX and Ola up to the 10 km mark, but the higher cost per km quickly multiplies beyond that point. And then we have UberBLACK and UberSUV which have the same rates but different capacities. They can actually offer a better deal than Meru and the likes for long distances over 25 km. Of course if you have 5-6 people travelling, then these 6 seaters are the way to go. Lastly, we have Ola’s version of the SUV with its Prime service that’s the costliest of the lot. Again, if you are in a group of 5-6 people, this can actually be cheaper than the taking two 4-seater vehicles, unless of course you manage to get a couple of UberGOs.

I haven’t considered the surge pricing in the above comparison, and that is a scenario where the older lot turns out to be cheaper. However, such scenarios are rare as Merus and the likes can be pretty hard to find for immediate travel. The interesting thing to see now will be the role that regulators play in toying around with these pricing models.

Update (16 Jun 2015): Found a major miscalculation in the trip time. I have corrected the graph and updated the text accordingly.