Anvehsan

Innovation musings

Week-end Musing: Why Do Indian Music Apps Show Hindi Film Songs in English Text?

July 5th, 2015

Have you wondered why all the great hindi film music on computers, mobile use English (phonetic) to present song titles ? Whether it is YouTube or Gaana or Spotify or any other service, it is common to see English script being used for hindi film songs.

As Gabbar would say, “बहुत बेइंसाफी है”. At Pariksha Labs, we think it is time we fixed this. It is time India adopted Hindi more widely on its computers and mobile. It is time Indian people moved to more natural language on computers and mobiles. Its time we included the non-english speakers, a large majority into the circle. Its time we #DesignedForIndia.

Stay tuned for some big news next week from our music app ‘Filmi Filmy”. क्यों, बात पसंद आई ?

Can Automated Text Summary Technology Work For News ?

July 2nd, 2015

 

Well, judge for yourself. The summary below is of a recent new story of a Pakistan trained doctor who has been forced to sell shoes in Ahmedabad. The summary has been produced by our automated engine that we plan to integrate with our News app, “inNews: The Big News Now”. Coming soon.

Summary

Pakistan trained doctor sells shoes in Ahmedabad

He fled to Ahmedabad in 2006 fearing for the safety of his family. Without the MCI’s permission to practice, these doctors often work in pharmacies or even mobile repair shops – all for a pittance. Most of the Pakistani doctors fled to Gujarat for safety. In Ahmedabad, he works at his cousin’s mobile store right next to the airport.

Do share your feedback.

Why Mobile News Must Break Away From Source->Category->Story Format

July 1st, 2015

 

Mobile users exhibit an entirely new way of consuming news stories, different from printed newspapers and even web-based news portals. Mobile users glance at their devices multiple times a day; they need to know the BIG, trending, breaking news and quickly see all the stories related to this news on the mobile device. They want the option to read detailed stories but only if there is significant interest. They are looking to share big, breaking news with their social and professional networks quickly to establish their own social brands.

Meeting this need requires a new way to present mobile news. The existing apps, both single brand news readers and/or news aggregator apps simply convert the printed/web style news to mobile instead of recognizing this fundamental difference for the mobile user. The result is that users must use multiple apps, visit multiple feeds and go thru multiple stories to identify the top stories and understand all points of view on the story. Even aggregator apps present news in the old catalog style: source->category->story. The user has to do all the hard work of browsing, move between many screens and takes a good amount of time to get a sense of what is truly trending.

In an age where computing is cheap, storage is cheaper and text processing is advanced, it makes no sense to follow old models for news presentation. News presentation must change and exploit the technological capabilities.

Check out our news reader, inNews (available on both App Store and Play Store) to get a fresh perspective on news reader. Why do the hard work of going thru 20 sites to figure out the BIG news when “there is an app to do that” :)

App Store: https://itunes.apple.com/in/app/innews-the-big-news-now/id991757311?mt=8

Play Store:https://play.google.com/store/apps/details?id=com.plabs.apps.innews&hl=en

 

From a Doodle to An App

June 21st, 2015

Here’s my whiteboard doodle that sort of captures how inNews works :)

Step 1: A crawler tracks news sources (RSS feeds) on a regular basis. We can go real-time but currently it is periodic to save on costs.

Step 2: The new feeds / stories are categorized, the headlines and the body of the story identified. The articles are archived in our vault. One day, we hope to be able to provide a “hor-news-grows” view and show the natural life-cycle of a  news story. A bit like the graphic below:

Step 3: Our keyword extraction kicks in and identifies relevant keywords in headlines, body, pictures.

Step 4: We use a clustering algorithm to identify related keywords and stories. Each story is matched to the cluster database. New clusters are created if needed.

Step 5: The clusters are sorted based on category, velocity and volume of stories

Step 6: The REST APIs use the cluster information to retrieve BIG news stories based on a data range; the # of stories related in the cluster and the summary of the story

Step 7: The app receives the response from the REST API and renders its UI. See a sample below:

What’s the BIG News Now?

June 18th, 2015

How do we know about big stories? In the age of newspapers, we assumed that the front-page headlines were big stories (and trusted the judgement of the editorial team). In the age of TV news, we assume that the stories featured in the 9 pm debates are the top stories for the day. When we look a breaking news thru the day on TV, we assume it must be a big news item to make it to breaking news.  In the internet age, we glanced at our favorite news web sites and saw the top headlines and assumed they were the big stories. But as social media, blogs and news portals have proliferated, we observed that different portals and channels prioritize different stories as their top stories – based on their judgement, bias and sometimes interest.

So how does find out what the top news of the moment is?

Our approach is algorithmic. We look at the volume of articles that are written on a theme across a wide range of sources. We co-relate related keywords into a thematic framework and determine which themes are getting popular. The top story is usually the one with the max number of interest, ie – individual sources reporting on that story.

This is the basis of our News Reader app – inNews. It uses sophisticated algorithms to track, process and analyze thousands of news feeds to simplify the top news for you. Open the app and Bingo – the top news of the hour is in front of you, with all of the related stories from all the sources so you can read as much as you like. Flip thru the top stories and in a quick 2 minute scan, you can see all the top 10-20 stories of the moment.

Or you can read thru 20 sites, and/or apps and do the heavy lifting yourself. Did I hear you say “Nah, that would be stupid in the age of mobile consumer with limited time?” We agree :)

So What are you waiting for? Grab the app and check it out yourself. The Play Store APK is live (https://play.google.com/store/apps/details?id=com.plabs.apps.innews&hl=en ). The App Store link should be up in a week.

 

Key learnings from building a “voice search” for hindi film songs

June 11th, 2015

A couple of posts ago, I talked about the idea that “audio search” makes so much sense for a music app. We have been working behind the scenes looking at voice to speech technologies and evaluating them with a view to offer voice search in our app “Filmi Filmy”.

We are happy to report that we were completely wrong when we first thought of this – Since all of the song titles are entered in English but represent Hindi words phonetically eg: “O mere dil ke chain”, “Gata rahe mera dil” we think that we can use a voice to speech engine to take user inputs, turn them into phonetic English and use the English text as the search keys.

It turns out that is it much more elegant and natural to take the voice input “O mere dil ke chain”, render it as the hindi string “ओ मेरे दिल के चैन” and search for the hindi string in the database. One significant advantage to this is that it reduces the complexity of the phonetics completely. It does not matter if the “ke” is spelled as “key” anymore as in Hindi it will always be spelled as “के”.

The challenge of course is getting a database of film song titles entered in Hindi. Nearly all song databases have English transliterated titles – and may we add- not two of them spell the same song the same way. A healthy inheritance from English led and US led software is that from YouTube to the home grown Gaana nearly all the songs are in English.

We are happy to report that fortunately a bit of innovation and tons of persistence can solve this problem (we may not have a huge cash chest at Pariksha but we are certainly not short on tech coolness). One of our engineers figured out a way to use existing open-source tools to build hindi equivalents of the titles.

The results are spectacular, to say the least. Consider for example this song search using voice search with hindi titles v/s text search with English phrases below:

Text Search With English Phrases Voice Search with Hindi Titles
   

 

We need to do a bit more work on the hindi song titles and improve the error handling on the search and this should be ready for public use. Now consider the scenario we had described earlier – Imagine slumping in a car after a long day and with no energy to type to search, all you have to do is say the song and voila the app will play it on your phone, ear-phone or connected blue-tooth speaker. Dare we say, it is not long before this will be a reality!

Building A Micro Payment Based Mobile Marketplace For Digital Content

April 23rd, 2015

The mobile phone has evolved from a communication device to a consumer device. Everything that we do in a typical day is moving to the phone. India has seen three primary drivers of mobile usage : voice & sms; messaging & social media; and now shopping.

The mobile usage trends show that not only is the traditional voice & sms category a near fully mature segment, even the demand for new-age applications are showing early maturity indicators.

 

At the same time, digital content consumption on the phone is witnessing a rising trend. While it has been hard for content owners, publishers and app developers to monetize this, early indications show that new-gen monetization models are beginning to have traction.

 

At Pariksha, we are building our vision of a micro-payment based mobile marketplace for digital content. We believe that as content consumption grows on the phone and as monetization models for content mature, there will be significant opportunity to deliver curated experiences for our customers through a portfolio of apps.

Here’s a sneak preview of our mobile marketplace for digital content and how our apps support this vision. I will write a detailed post on the marketplace next week. Meanwhile, if you have any ideas/suggestions, feel free to send to info@parikshalabs.com.

 

 

Why “Audio” search makes so much sense for a music app

April 13th, 2015

Our app “Filmi Filmy” provides users with a fast and simple way to view their favorite Hindi film songs. We have observed that YouTube has a big collection of Hindi film song clips but the lack of phonetic search and presence of multiple clips with same title intimidate users. “Filmi FIlmy” provides a friendly Hindi phonetic search so that we can find the songs you want irrespective of how the titles are spelled. Also, since we only pick up a single entry for any song, the user is assured of getting the right song clip without having to sift through multiple of choices. The value of “Filmi Filmy” can be seen in the use-case metrics – over 6.5 minutes per session engagement and over 40% repeat users from all the “Indian diaspora” hot-spots on the globe bear testimony.

At Pariksha we are constantly asking ourselves – how can we use technologies and make it easier for the users to do what they want to do. How can we make it faster, more accurate and more user-friendly. For now, our thoughts are locked in on “voice search”. The popularity of SIRI and voice input on android platform point to increased user convenience especially when used in a limited context.

Since all of the song titles are entered in English but represent Hindi words phonetically eg: “O mere dil ke chain”, “Gata rahe mera dil” we think that we can use a voice to speech engine to take user inputs, turn them into phonetic English and use the English text as the search keys. The user will get the feel that the app is searching for the songs basis his/her voice. Our look-ahead mechanism will show users all the songs matching the phonetic tokens and in the case of any default, the user can type out the search pattern directly.

Imagine slumping in a car after a long day and with no energy to type to search, all you have to do is say the song and voila the app will play it on your phone, ear-phone or connected blue-tooth speaker. You can drive hands-free! Imagine being able to do this at home and flick the video to your large TV (thru the chrome-cast interface) – ultimate song viewing for the aficionado.

With this in mind, we are starting an initiative to integrate “Filmi Filmy” with the best speech to text technology. Watch this space for an exciting new innovation that you will enjoy, appreciate and other music apps will emulate :)

 

 

 

Does Blocking A Content App Restrict Innovation? We Think It Does.

April 6th, 2015

When we launched “Filmi Filmy”, our app to bring Hindi film song videos from YouTube into a curated experience last year on iOS platform, our hunch about the latent user demand for such an app was proven immediately. We raked up 30,000 downloads in a month and were rated as the #1 new music app in India and #1 music app in Pakistan. The usage stats were awesome, 50% plus repeat visitors and over 8 minutes engagement per session. Indian diaspora from all over the world were searching, playing and enjoying the new found ability to search for Hindi film songs phonetically and enjoy clutter free viewing experience.

At Pariksha, we felt very energetic and enthusiastic. We had demonstrated that a) there was appetite for a Hindi film videos app b) users prefer the high success rate of phonetic search and c) users want convenience and simplicity and not irrelevant choices and having to make endless decisions.

With this in mind we invested further into the product. To expand our reach, we started working on an Android version and even a Windows mobile version of the app. To experiment with business models, we built an in-app purchase mechanism to allow users to save and share playlists. We would allow users to buy credit points and use that to store upto 10 songs in a playlist. Saved playlists could be retrieved from any device and could be forwarded to a friend/family member. If you wanted to impress your girl-friend and praise her eyes, you could send her a playlist with your favorite “nayan” or “nain” or “naina” songs! Playlist could be used to express feelings – just like emoticons. It was an innovation experiment to see if playlist curation could be monetized.

The launch of “Filmi Filmy” on android was very successful. We had 5000 downloads within a week. However, what happened next was a huge dampener. Google blocked the app without any specific reason, claiming we had violated some terms of usage. We requested to know the specific violations and were ready to revise the app to comply with terms of use but in Google land, one deals with an automated system that churns out standard responses with no intelligence to interact with. We were well and truly stumped and out of the largest opportunity – the great Indian android market. We were bumped off the Play Store and facing a huge loss of investment and opportunity.

Interestingly, Google did not complain to Apple Store about the iOS app. It seemed that they were using their control over Play Store to block the Android app but were not sure about the infringement themselves to worry about taking it up on Apple Store.

The block forced us to revisit our strategy. We pulled back promotions and slowed development to try and figure out a new approach. I even tried to approach YouTube team for a potential partnership – after all, we had innovated a new way to use the video content and who best to take an innovation to the masses than a big partner. No luck here too, YouTube is happy to partner if you are a content owner but if you have better ideas than them on using the content, they don’t want to know about it.

Over the next 6 months, we let the iOS app move on its own steam. We were happy to see regular download volume, consistently high engagement metrics and a fairly regular in-app purchase behavior.

We feel that blocking our app on Android restricted our ability to test the key innovations in the large Indian market. It damaged our ability to establish playlist sharing as another “eMoji” type and restricted users from getting access to a simpler, cleaner way to see Hindi film song videos.

It also restricted YouTube’s own ability to monetize its assets. If more people used “Filmi Filmy” and we could work out a partnership to share in the monetization, everybody would benefit. Blocking the app was a loose-loose proposition while it could have been a win-win deal.

But does anyone at Google/ YouTube care? Will they care?

(Check out “Filmi Filmy” and its innovations. Download the app on iOS and Windows and let us know if you agree it provides a better experience for Hindi film song videos than any other music app out there)

Can Blendle style pay-per-article work for News in India? Or is it too early?

March 30th, 2015

Over the weekend, I cam across Blendle , a dutch start-up that has built a news aggregation platform that allows users to read on a pay-per-article model. It has received early recognition and some financial support from the top publishing companies, such as New York Times and Axel Springer. It has also got the digital media world sit up and notice – after all with decreasing print revenues and increasing competition from digital media companies, aggregators and click-bait savvy upstarts, the traditional media companies are having a hard time monetize their content.

As someone with long standing interest in digital content – formats, creative tools, distribution tools and of course business models, AND a portfolio of apps that are trying to find our own answers to the monetization puzzle, I found this very interesting.

What if we took our news reader, inNews to a similar model?

We already use a virtual credit point model in our apps. In fact our credit points can be used across apps. Users can re-charge their virtual credit points through an in-app purchase of credits (from the store) and use it for transactions in our apps. Filmi Filmy (our hindi film songs app) allows users to store their playlists and share their playlists with others using the credit points. ThotOfU (our custom greeting card maker app) uses the credit points to purchase interesting pictures. We have demonstrated, although at a very very small scale that users are willing to buy credits and transact for these services.

What if we could offer the user seamless reading experience for detailed news stories against their virtual credits ? We could think of our credit points operating like a virtual currency for a digital content marketplace where you can buy pictures, storage, playlists, news and perhaps in the future any digital content for education, information or entertainment.

Its an intriguing thought. I will be discussing this with our developers and product team. If you have any ideas please feel free to share your ideas. I am painfully aware that India is a late bloomer in tech adoption, attitude changes and anything related to payments in India has a huge resistance. Yet, as someone who has always been at the leading edge of risk and ideas, I prefer to die by innovation than die without trying.

Blog

Its an intriguing thought. I will be discussing this with our developers and product team. If you have any ideas please feel free to share your ideas.

I am painfully aware that India is a late bloomer in tech adoption, attitude changes and anything related to payments in India has a huge resistance. Yet, as someone who has always been at the leading edge of risk and ideas, I prefer to die by innovation than die without trying.

More News

Its an intriguing thought. I will be discussing this with our developers and product team. If you have any ideas please feel free to share your ideas.

I am painfully aware that India is a late bloomer in tech adoption, attitude changes and anything related to payments in India has a huge resistance. Yet, as someone who has always been at the leading edge of risk and ideas, I prefer to die by innovation than die without trying.

Links

Its an intriguing thought. I will be discussing this with our developers and product team. If you have any ideas please feel free to share your ideas.

I am painfully aware that India is a late bloomer in tech adoption, attitude changes and anything related to payments in India has a huge resistance. Yet, as someone who has always been at the leading edge of risk and ideas, I prefer to die by innovation than die without trying.