Lattice Bridge: breaking the communication barriers with its Speech Recognition deftness
When Apple introduced its soulmate Siri, the world reverted with oddity. Until then having a conversation with your mobile phones or computers seemed pure science fiction. Today, with voice-controlled interfaces showing up in mobile phones, TVs, and automobiles gives us a clear vision- it’s all about the voice. As homo sapiens, we have found our new talk-to-companion via our smartphones, TVs and others, asking them to cover for us in sending e-mail and text messages, playing our favourite playlist, searching for directions, putting our fingers to rest!
A report presented by Grand View Research, Inc., predicts the global speech and voice recognition market size to reach USD 31.82 billion by 2025. As we are moving towards the advancement, the demand for voice-activated systems, voice-enabled devices, and voice-enabled virtual assistant systems, IVRS is mounting, owning much of its progress to technology-backed banking and automobile sectors. While safety and security are accelerating the adoption of voice biometrics, AI and Machine Learning are soaring the need to understand human speech. But before that, we are left with much more to do with human beings too.
India, a land of vivid wonders, language and diversity, also nestles a population of almost 50-60 per cent out of the way of the digital revolution. Understanding the ABCs of mobile devices to connect with the world was definitely a hurricane, fuelled by lack of connectivity and English literacy. Until a path was set in 2002 to connect this side-lined population with the mainstreams of banking, healthcare, education and other sectors.
Growth Story So Far
Today, walking on the footsteps of iPhone’s Siri Technology’s success, Amazon’s Alexa and many others, Indian start-ups are betting on speech recognition technologies. But one person created the path far ahead before Apple or Google could. Speech-based technologies hold a history of almost two decades, but with the advent of ‘personal assistants’ like Cortana, Siri, Alexa, Google Home, we have now opened the gates for this technology, the ‘voice-era’, powered by AI and Machine Learning.
C Mohan Ram drew the outline of a vision to empower people to interact and transact with diverse information systems using Speech Recognition Technology called ‘simply speak’, though LatticeBridge Infotech, one of the early pioneers of Speech Recognition Technology. So, what was the hurdle that led to the inception of LatticeBridge Infotech?
Our country’s diverse language portfolio doesn’t come handy when training speech engines. Natural Language systems’ abilities fail whenever they are faced with dialect based on multiple languages, thus presently inaccuracy in catching things and that’s why LatticeBridge’s deployments are less in the natural language processing mode, and more on the directed dialog mode, where the caller is led through a series of Yes/No type questions. Today, the frontrunner’s esteemed clientele includes Indian railways and leading banks and telecom companies. Apart from 11 Indian languages, the company supports Arabic, which has helped the firm bag clients such as major telcos in the UAE and National Airlines company, as LB customers.
Emerging as a differentiator
LatticeBridge (LBIT), a pioneer in Speech Technology space empowers people to communicate with information systems using voice commands. LBIT’s cutting edge speech technology & tools helps in delivering high-value, cost effective information solutions in record time. At LBIT, People, Product and Processes converge to provide the competitive advantage that upwardly mobile organizations need today to succeed in their business.
Here is an excerpt from Mr Mohan’s previews interviews, “Even in our case where we’ve successfully deployed and shown very good live scenarios handling millions of calls, we’ve had to remove it (natural language processing) because of mixed languages. If you remove AI or learning, then the system doesn’t deliver on an optimal percentage of automation, and then the customer gets disappointed,” says Ram. LatticeBridge’s deep learning speech engines have “failed miserably” in Kannada especially, with 77% accuracy, he says, while in Hindi and Tamil, it is 98%.
Not all forms of speech-based AI are the same, and aren’t comparable by the same yardstick, says Mohan. He classifies speech AI systems into four categories:
- embedded systems such as those in a car, with a limited vocabulary of commands,
- desktop systems, which have a slightly larger vocabulary
- dictaphone type software where you have to train the system with your voice, and
- networked speech applications, which are designed to support millions of users.
We (LatticeBridge) are positioned in the network speech application space and we are not doing the other three,” he says. “In this space, there are no limitations in terms of processing power, which means that Gaussian mixture models and Hidden Markov Models can be loaded, and intelligent deep learning algorithms can be linked there, as Millions of users speaks Billions of words in Trillions of manners!
- Automatic Speech Recognition (ASR) based customer service solution
- Voice Biometrics based authentication & fraudster detection in Contact Centre
- Versatile Multi-Channel, Multi-location, Multi-Lingual – Contact Centre solution
- Consulting Services for “Customer Relationship Management”
LBIT- TELE SUITE
Simplicity Packaged with Desired Scalability & Ultimate Reliability. LBIT’s “LB-360” suite is simple to use but still provides the flexibility to integrate all Business logic with ultimate reliability. LB-360 comprises “Building Blocks” of pre-packaged telephony features that facilitate easy and rapid development of solutions for “Customer Interactions Management”. The endless lists of solutions are; Speech‐enabled Solutions, Contact Centers, Predictive Dialers, Automatic Voice Response Systems, Unified Messaging, and Automatic Information Delivery and voice biometrics for authentication.
LBIT follows the tested and proven FIND development methodology. FIND is requirements-driven and architecture centric. It follows a 4 stage structured, progressive and iterative approach starting with an in-depth ‘Feasibility’ study. The ‘Initialize’ stage is where the Prototyping and Application Architecture takes shape. Highly skilled developers with more than 5 man-years experience in speech application development handle the ‘Nurture’ stage, which involves design & development, and integration testing. As part of the ‘Deployment’ stage, multi-cycle testing is done internally to ensure defect-free, faster and easier UAT.
Go after Global market. We have recently replaced Huawei customer interaction system in a large telecom company in UK. Now, with a ban of this company in USA, we see more opportunities to grab the overseas market. N.American and European markets are going to be our focus for next 2 years. In terms of product improvement, integration with AI/ML based predictive recognition for “proactive care” (instead of customer calling for information, enterprises know the customer pattern of interaction and serve it proactive way) as well as behaviour analytics is our goal to serve customers using latest technologies. Also, moving the entire offering on ‘Cloud’ is being worked upon by LB team.
Won the “Most Innovative company in CRM” from NASSCOM in 2009.
An up-close with the brain behind the brand: C Mohan Ram
M.Tech from IIT-Roorkee (passed in 1986)
- CAREER BREAKTHROUGH
Shift from technology role as X-Windows specialist to Sales and Marketing in 1993, covering Oil, Government and Utilities sector across Globe.
Digital Signal Processing with several innovation in enhancement of Signal to Noise Ratio to detect the low volume signals / speech inputs. Now specialising on Cyber Security education and research, with focus on Cyber-Physical Systems Security.
- INSPIRED BY
Many including Vivekananda to Abdul Kalam. But my main inspiration is my father, who taught me the fundamentals of living and English. He stayed through the struggle of learning and corrected my mistakes patiently to help me tune the same to achieve where I am … His famous quote has been my inspiring fuel, “Be prepared for the worst and hope for the best. A shade better than worst will make you happy”.
- BIG WINS
Sold a USD 6 Million project to a large oil company using my techno-commercial skills for ‘Year 2000 migration’ for that company.
- YOUR TRADEMARK
Whatever I start, I finish with best of my abilities before I sleep … as I am a perfectionist, it is difficult for people working under me to satisfy me!
- THE JOURNEY
Having moved from a comfortable month-end salary + sales incentives and annual bonus with vacation etc., it is an ocean of change in our life style after starting the business. I realise that doing business without deep pockets is very difficult. But, God has been kind to give enough and more opportunities to learn many things in my life. Most importantly, Patience is virtue of any success. Despite seeing several ups and downs, which is typical of any business, being patient without losing hope has made us survive the toughest of toughest moments in running the business. Customers are happy and satisfied, as many of them are now customers for more than 14 years out of 16 years of existence, as we strive for ‘customer delight’ in all terms, without minding the money earned or profitability. Now with overseas market available for us, we are certain to reach our dream of being a Unicorn company in next 3-4 years.
- PIECE OF ADVICE
Young entrepreneurs think VC money is Easy money. However, it is not so. Equity is the most expensive form of financing and comes with a lot of strings attached to it. Try to boot-strap your operations to reach a position that you are in a position to bargain for the deal you wish. Never take money out of desperation! Also, now a days we find that youngsters are not giving priority to health and exercise. No matter what the you get up, walk for 45 minutes at least! That’s my two cents advice.… to Next Generation.