Posted on Dec 03, 2018

Compliance Monitoring enters AI age at ING with Intelligent Voice and Relativity Trace

Find the PR Newswire Release here       Chicago, December 3rd, 2018 – Compliance monitoring has become a hot topic in artificial intelligence (AI) in 2018. This has led a number of banks to explore how new technology can help reduce fraud, money laundering, and insider trading amongst other regulatory infractions. ING Bank started […]

Find the PR Newswire Release here

 

 

 

Chicago, December 3rd, 2018 – Compliance monitoring has become a hot topic in artificial intelligence (AI) in 2018. This has led a number of banks to explore how new technology can help reduce fraud, money laundering, and insider trading amongst other regulatory infractions.

ING Bank started this journey early; in 2015 they engaged with Intelligent Voice®, a leading supplier of speech recognition and compliance software in the financial services market, to see how innovative technology can be used to improve their processes.

During that time, Relativity, the industry-leaders in e-Discovery software, started its partnership with Intelligent Voice to deliver an integrated, end-to-end system that would bring together the best of Intelligent Voice’s AI technology with Relativity’s world class platform. Now, ING has signed on to leverage Relativity Trace, a new app for proactive risk management and compliance. ING plans to use the combined solution to proactively monitor and stop suspicious activity, like insider trading, collusion, and improper investment management practices, before they become a problem within its business.

ING has gone live with Intelligent Voice and Relativity Trace in its Amsterdam location and is already monitoring almost 400 traders across email and IM, with voice following shortly. This is the beginning of a global roll out that will monitor data from over 1200 traders worldwide in a dozen languages.

 

“The combination of IV and Relativity has allowed us to build a fully distributed review capability for the compliance monitoring team capable of handling all electronic communications. The small footprint offered by IV’s GPU powered speech recognition means we can deploy this at a significantly lower operating cost than other solutions saving both rack space and energy. Next to that we have seen that the voice transcribing accuracy for our purpose is proven higher that 85%. Installation started in July 2018 and we are already up and running with the first phase in November 2018.”

Paul Braakman, ING

 

It has been our pleasure to work with ING in bringing this innovative new package to market. The professionalism of the ING team has allowed us to give the bank great value in a short period of time.

Ben Shellie, CEO, Intelligent Voice

 

We are thrilled to partner with Intelligent Voice and ING on this exciting initiative. Intelligent Voice’s connectors enable ING to push all of their communication data proactively to Relativity Trace, so their compliance team can quickly take action on suspicious activity.

Jordan Domash, General Manager of Relativity Trace

 

IV and Relativity Trace are available as both on prem and cloud solutions to enable companies large and small to monitor all electronic and audio communications.

Intelligent Voice coupled with Relativity provides the industry’s most capable and secure audio, email, and chat review platform.

 

About Intelligent Voice®

Intelligent Voice Limited is a global leader in the development of proactive compliance and eDiscovery technology solutions for voice, video and other media. Its clients include government agencies, banks, securities firms, Call-Centers, litigation support providers, international consultancy, advisory businesses and insurers, all involved in the management of risk and meeting of multi-jurisdictional regulation.

Fundamental to its success, its patent-pending and patented technologies Intelligent Voice and JumpTo™ are developed by a team of dedicated researchers and system engineers based in the UK. Ownership of the core technology resides with Intelligent Voice. Intelligent Voice continues to lead the market and will maintain its strengths in the areas of thought leadership, innovation, R&D and providing solutions to its clients.

The system uses speech recognition technology to capture calls, convert them into text and then automatically send the transcript (along with the original voice file) to the user’s inbox, as well as provide complex analytic capabilities

The Company is headquartered in the United Kingdom and has been providing market leading solutions since 1990.

For further information about Intelligent Voice, please visit https://www.intelligentvoice.com

Contact: Jessica Harvey

Email: [email protected]

Tel: +44 203 627 2670

 

About Relativity

At Relativity, we make software to help users organize data, discover the truth, and act on it. Our e-discovery platform is used by thousands of organizations around the world to manage large volumes of data and quickly identify key issues during litigation, internal investigations, and compliance projects. Relativity has over 180,000 users in 40+ countries from organizations including the U.S. Department of Justice, more than 70 Fortune 100 companies, and 198 of the Am Law 200. RelativityOne offers all the functionality of Relativity in a secure and comprehensive SaaS product. Relativity has been named one of Chicago’s Top Workplaces by the Chicago Tribune for seven consecutive years. Please contact Relativity at [email protected] or visit http://www.relativity.com for more information.

Contact: Taylor Laabs

Email: [email protected]

Cell: 952-261-2690

SOURCE: Relativity

Posted on Nov 27, 2018

Is Speech Recognition just OCR for Voice? I don’t think so…

Is Speech Recognition just OCR for Voice?  I don’t think so… In 2015, I put up a slide at a Financial Services technology conference saying “2015 is the Year of Voice”, citing examples like Siri and the massive fines then being handed out for rogue trading as reasons why the technology was being taken seriously.  […]

Is Speech Recognition just OCR for Voice?  I don’t think so…

In 2015, I put up a slide at a Financial Services technology conference saying “2015 is the Year of Voice”, citing examples like Siri and the massive fines then being handed out for rogue trading as reasons why the technology was being taken seriously.  The following year it was “2016 is the Year of Voice”.  And then 2017 rolled on, as did my slide.

Finally, I’m sure, 2018 is the Year of Voice: I think.

Alexa more than any other device has shown the potential for voice technology.  For the first time, someone has put into production a solution that is actually usable by consumers, that doesn’t require pre-training, and does something useful.  Anyone who has tried to use transcription software (which was the last real attempt at using voice technology to do something to make people’s lives easier), knows that even small imprecisions in the output made it quicker to learn to type.  Finally, you can walk into your kitchen and turn on the radio just by asking.  Progress!  Who needs pesky buttons.

“Alexa … has shown the potential for voice technology”

But it has highlighted some more interesting issues, like how do we feel about having an always on microphone in the house, and how do we feel about all of our sensitive data passing through Amazon’s hand.  I have often cited the “Magic Pipe” as one of the great dangers of current implementations.

When you use a web browser to contact your bank, the whole transaction is secured end to end using SSL.  But when you say “Alexa, what is my bank balance?”, those voice commands are sent to Amazon, who send the question to your bank, who send the information back to Amazon who send it to your Alexa device.  There is no Magic Pipe of secure data connection between your somewhat underpowered Alexa device and the bank. How comfortable do you feel now?

While Privacy is undoubtedly an issue with home voice devices, as it is with using any public cloud service, what I want to explore are the common mistakes that people make when assessing and implementing speech recognition technology.

90% accurate, but 0% useful

Speech recognition is hard, and it is complicated by a lot of factors.  Telephony speech in particular, is hard as it is highly compressed and often conversational.  When you then add in multiple parties on a conference call, it can be hard even as a human to really grasp what is being said.  Have you ever tried to record a meeting using the recording app on your phone?  It is very easy to strategically place your phone so that no-one comes out clearly, a combination of the “far-field” effect of distance from the microphone and background noise.

“Have you ever tried to record a meeting using the recording app on your phone?”

It is one thing to accurately transcribe clearly spoken speech, but quite something else to transcribe things accurately in the wild.

The first question we get asked when talking to customers is “How accurate is your speech recognition engine?”.  There are all sorts of answers to that.  “As good as Google” is one. “Better than IBM” is another.  Although true, none of these replies are that helpful.  I could say “95% accurate” or “50% accurate” because both of those would be true as well.  It is dependant on the use case.  So, our answer to that question is another question or often a series of questions, the primary one being “What do you want to use speech technology for?”, and the answer is usually “We want a transcript”

Let me tell you very clearly. In almost every use case involving speech, the one thing you don’t want is a transcript, at least not in isolation.

This leads on to what our first question should probably be, which is “Why do you think you want a transcript?”

There are a number of answers to this.  Some are for search use cases, where people want to retrieve files at a later date using text search.  Others are for keyword spotting (compliance, QA and fraud often need this).  Others are NLU, where they are interested in the intent.  Some will be for sentiment analysis.  And the worst one “We want to read and review the transcript”

Have you ever read a transcript, particularly of a telephone call?

Try this little snippet for size:

Believe it or not, this is between two people at the top of the “articulate” scale, one of whom will probably be the next King of England (so I have to be careful what I say, lest my head be severed from my body).

The point is that conversations are not like properly written documents (like emails, for example).  They are difficult to “read” even when they are perfectly transcribed.  And the reason why there is still a thriving “human in the loop” transcription industry (costing $60+ per audio hour for the transcript) here in the Age of Alexa is that machine transcripts are still not that accurate, especially in the more challenging environments identified above.

So, if speech technology is not 100% perfect “OCR for Voice”, what is it and can it do?  And how can it be used for better business benefit?

Training

You wouldn’t set off on a 100k bike ride without doing a little training?  Well don’t start your voice project without it either.  We used to be fairly soft with clients about this.  We would say “We do have a process that quickly allows you to improve the recognition accuracy, but feel free to run it out of the box if you want”.

But we found that in every case the customer didn’t get what they wanted.  This is because every domain is different, and unless you adapt to that domain, you get good general results, but you don’t get results that match your expectations.  This is often noticeable where people have a search or keyword spotting scenario.  If am on the lookout for a particular unusual name or a person or product, it is almost certain that a speech recognition system will not have it in its “lexicon” (a dictionary to you and me) of words.

Want to catch references to “Martin Shkreli” in your voice data?  You won’t unless you search for “Martin Squirly” or unless you have pointed your speech recognition system to the Turing Pharmaceuticals Wikipedia page (here, if you are interested), and said “Learn this”.

Ask Alexa to play “The Word Girl” by Scritti Politti.  She’ll get it (unfortunately, from a musical standpoint), because “Play” sends her off to her dictionary of musical words and terms, and so she’s pre-sensitised to artist, song and album names.

Not magic, just domain-specific training.

Seeing the Future

If you speak into a “live” speech recogniser like Siri, you will often see phrases appearing before you have said them, and you may well see words that have been transcribed change, often from something wrong to something right.  This is because speech recognisers use statistical models to help them work out what humans “mean” when they are speaking to try to get the right words.

Start a sentence “The cat sat on the”, you listener (and your speech recognition engine), is expecting you to say “mat” at the end.  Say “rat” and you have engendered a state of confusion.

This power to see the future also helps in certain other scenarios.  When we do machine transcription, we only expect to see single lines of text as output, representing what has been said.  But in the background, where the system is not necessarily sure what you meant, it will have generated alternatives, often homophones of what it heard (Claus, clause and claws for examples).

If your use case is trying to find things, like for search or keyword spotting, you really want to have access to that data, because it will expand your search pool.

This is called a lattice, a list of alternative words with a “confidence” score attached to it.

Below you see an example of a phone call.  Every single word that was said was correctly identified, so if you had used the lattice to search, you would have found what you were looking for.  Especially as the name, often a key search term, is only seen with a confidence of .08.

Is Voice Ready for Primetime?

Yes, and no.  And whether it is very much depends on you.

Step 1, understand your use case.

Step 2, really understand your use case.  Unless you are trying to produce close captions for a podcast, you don’t just want a transcript, “out of the box”.

Here are some questions to ask:

Is my use case really a “search” use case?  If so, think about how you use training and lattice output.  Microsoft did some research on recall using lattice results and found:

“Experiments on a 170-hour lecture set show an accuracy improvement by 30-60% for phrase searches and by 130% for two-term AND queries, compared to indexing linear text.”

This means that if you search for a phrase, say “Intelligent Voice”, you are up to 60% more likely to pick it up a lattice search than if you search the plain text output you usually see from a straight transcription. If you make that a Boolean search “Intelligent AND Voice”, the chances of picking it up rise to 130%

Am I worried about privacy? Are you happy that your data is commercially or legally suitable to be processed by a public cloud provider?  If not, you need to look at on-device, on-prem or private cloud solutions

Am I worried about cost? Google charges $1.44 per audio hour processed.  Think about the volume of data you might be processing and multiply it by that.  If you are looking at telephony monitoring solutions with agents on the phone 30 hours a week, does that make commercial sense?

The abilities of voice technology do not really live up to the hype.  Much commercial speech recognition is focused heavily on English, and “normal” accents.  And they usually rely on the audio being processed being pretty clean to start with: Alexa uses a 7 microphone array and clever “beamforming” techniques to overcome the problem of you being distant from the microphone.

At the moment, we are building a model for the Bavarian dialect:  not Bavarian accented German, but an actual dialect, but reduced in written form to Standard German. Not an easy task, but another step in trying to break the Anglo hold on speech recognition.

Will it ever be 100% accurate as well as 100% useful?  As they say in Munich, “Schau ma moi”.

 

 

Posted on Nov 08, 2018

IV Connect™ for Relativity Trace

Intelligent Voice® – IV Connect™ for Relativity Trace London, 8th November 2018 Relativity this week announced the launch of Relativity Trace, a new application built on the Relativity platform that  provides businesses with the tools to proactively monitor internal communications and flag the highest-risk content for further review. Trace can automatically ingest all forms of communication […]

Intelligent Voice® – IV Connect™ for Relativity Trace

London, 8th November 2018

Relativity this week announced the launch of Relativity Trace, a new application built on the Relativity platform that  provides businesses with the tools to proactively monitor internal communications and flag the highest-risk content for further review.

Trace can automatically ingest all forms of communication from your systems. IV Connect for Relativity Trace provides the connection from these systems, including email, audio and chat, to Trace’s ingestion capability.

Intelligent Voice’s IV Connect platform encompasses more than 20 proprietary systems used by trading platforms across telephone, email and chat, allowing high speed and high-quality data to be passed directly into archiving and review systems.

IV Connect

 

 

 

 

 

 

 

 

 

Intelligent Voice® installed its first pioneering holistic compliance monitoring system in 2012 and so has led the way in providing sophisticated solutions for gathering and analysing audio and other data, allowing compliance officers to do their job quicker and more effectively.

 

With the increase in the adoption of AI driven compliance, particularly in highly regulated organisations, having the ability to quickly and easily identify suspicious activity should be at the top everyone’s priority list.

Organisations can seamlessly integrate their core voice and text-based platforms into Relativity for real time monitoring, meaning all your compliance team has to worry about is reviewing alerts – not manipulating and moving data.

Intelligent Voice provides compliance, review and monitoring systems to more than 40 companies across the globe, including a number of the world’s major banks.

We are pleased to say that a number of early adopters of Relativity Trace are existing Intelligent Voice clients and are now already using or in the process of introducing IV Connect for Relativity Trace as a solution to their need to simultaneously monitor all communications, across multiple trader bases, across more than a dozen different languages.

Relativity Trace allows sophisticated analysis, reporting and review, particularly for large scale multi-country and multi-functional teams. Using IV Connect, organisations can detect and act on suspicious activity, including insider trading, collusion, improper investment management practices, bribery, and other high-risk scenarios.

 

“The ability to capture multiple data sources into Relativity Trace, especially audio, is essential for companies looking to build a robust and scalable compliance monitoring platform, both in financial services and other areas. We are delighted that Intelligent Voice has integrated its IV Connect product into the Relativity pipeline. “

Jordan Domash, Relativity

 

Intelligent Voice lead the way in its use of GPU technology for audio and data analysis. This technology is the driving force behind Intelligent Voice’s significant product portfolio, engineered to support client requirements, particularly for large enterprise installations.  Some examples include:

  • Biometric authentication and search, to allow for individuals to be identified by voice and not just metadata.
  • Encrypted Search: the ability for data to be encrypted on-premise, but to be searchable in the cloud using encrypted search terms. This gives clients the ability to safely move data to the cloud, while allowing for all of the security of an on-premise installation.
  • Real-time web-based transcription: Webex and other web-based chat voice can be difficult to capture. Intelligent Voice has developed a system to allow for this audio to be captured and stored in real-time. This also allows for real-time review of transcription quality.
  • Image recognition: Intelligent Voice has significant experience in the use of AI technology for image recognition – Recent examples of this include developing a crash damage assessment system for a major insurance provider.

Intelligent Voice is constantly looking at new features and innovations that might help its clients’ businesses. Together, Relativity and Intelligent Voice already provide the most capable and secure audio review platform in the industry.

 

About Intelligent Voice®
Intelligent Voice Limited is a global leader in the development of proactive compliance and eDiscovery technology solutions for voice, video and other media. Its clients include government agencies, banks, securities firms, Call-Centers, litigation support providers, international consultancy, advisory businesses and insurers, all involved in the management of risk and meeting of multi-jurisdictional regulation.

Fundamental to its success, its patent-pending and patented technologies Intelligent Voice and JumpTo™ are developed by a team of dedicated researchers and system engineers based in the UK. Intelligent Voice continues to lead the market and will maintain its strengths in the areas of thought leadership, innovation, R&D and providing solutions to its clients.

The system uses speech recognition technology to capture calls, convert them into text and then automatically send the transcript (along with the original voice file) to the user’s inbox, as well as provide complex analytic capabilities

The Company is headquartered in the United Kingdom and has been providing market leading solutions since 1990.

For further information about Intelligent Voice, please visit https://www.intelligentvoice.com

 

About Relativity
At Relativity, we make software to help users organize data, discover the truth, and act on it. Our e-discovery platform is used by thousands of organizations around the world to manage large volumes of data and quickly identify key issues during litigation, internal investigations, and compliance projects.

Relativity has over 180,000 users in 40+ countries from organizations including the U.S. Department of Justice, more than 70 Fortune 100 companies, and 198 of the Am Law 200. RelativityOne offers all the functionality of Relativity in a secure and comprehensive SaaS product.

Relativity has been named one of Chicago’s Top Workplaces by the Chicago Tribune for seven consecutive years. Please contact Relativity at [email protected] or visit http://www.relativity.com for more information.

 

Posted on Oct 16, 2018

How Artificial is your Intelligence?

One of the big factors holding back the adoption of Artificial Intelligence in technology in sensitive fields is it lacks “explainability”. Many people find relying on “black box ” technology too risky to make potential life and death decisions. This is operated across many verticals including Government, Financial and Healthcare where we are increasingly using […]

AI

One of the big factors holding back the adoption of Artificial Intelligence in technology in sensitive fields is it lacks “explainability”. Many people find relying on “black box ” technology too risky to make potential life and death decisions. This is operated across many verticals including Government, Financial and Healthcare where we are increasingly using advance machine learning algorithms to provide diagnosis that cannot be seen by humans.

 

GPU Accelerated Speech Processing

Intelligent Voice® is the world leader in the provision of GPU accelerated speech recognition, having first released GPU power in 2014, and is now established across the globe processing tens of millions of hours of audio every year across a wide variety of verticals including Government, Financial, Healthcare and Prisons. What is less known is that Intelligent Voice has a leading machine learning practise that assists its customers in building new processes for their businesses using AI. This started with pioneering work with the use of spectrograms to produce more accurate speech recognition (For an example have a look back on Nigel Cannings talk at GTC 2016 on Deep Convolution Neural Networks – http://on-demand.gputechconf.com/gtc/2016/presentation/s6371-nigel-cannings-deep-convolution-neural-networks.pdf ) and has now widened out to provide assistance to companies such as call centres and insurers.

 

Explainability

 

The advent of GDPR in Europe highlighted one critical problem with the use of AI for decision making; you can no longer use “computer says no” as a reason for automated decision making e.g. declining insurance. GDPR is beginning to force us towards a doctrine of “explainability” how the computer reached the decision that it made.

Anyone familiar with machine learning technology knows two things:

One – under the right circumstances it is easy to spoof, see below examples:

Two – that it is extremely difficult to provide an explanation of how a decision has been reached.

 

“Human in the Loop”

 

One of the themes we will be exploring at GTC DC this year is how to build networks that are able to provide a path through their decision making to allow you as recipient of intelligence, the ability to double check and potentially cross correlate the intelligence you have been given.  We see AI as the beginning of a revolution that enables humans to make significantly more decisions across a wider range of data, significantly more quickly; “human in the loop” is a vital step in the final decision making of many sensitive organisations. We have all heard tales of how AI will take us to a world of SkyNet; whilst that may be the stuff of sci fi, in reality we still want to understand how machines work.

 

#GTC18

Join Intelligent Voice’s Nigel Cannings on Tuesday 23rd October in the Monroe suite for his talk “How Artificial is Your Intelligence?”

Not yet registered for GTC DC? Use code GMCGTC18 for 20% off registration. Government attendees can register for free.

Posted on Feb 27, 2018

Castel/Intelligent Voice process 15 million hours of audio every year.

Castel Detect™, Castel’s call monitoring software, is successfully monitoring 15 million hours of telephone calls every year using its proprietary on-premise solution. Today it is announcing the availability of the same rich functionality as a cloud-based offering. Powered by Intelligent Voice®, Castel Detect™ delivers fast and accurate word and phrase detection for customer/agent conversation monitoring […]

Castel Detect™, Castel’s call monitoring software, is successfully monitoring 15 million hours of telephone calls every year using its proprietary on-premise solution. Today it is announcing the availability of the same rich functionality as a cloud-based offering.

Powered by Intelligent Voice®, Castel Detect™ delivers fast and accurate word and phrase detection for customer/agent conversation monitoring across a wide variety of industries. Telephone call compliance and monitoring is becoming increasingly important across a wide range of industries, from call centres to law enforcement and prisons.

Intelligent Voice’s speech recognition engine based around NVIDIA® GPU technology, leverages the massively parallel world of CUDA programming to give blisteringly fast ASR (Automatic Speech Recognition) across large data sets.

Our continued partnership with Castel, dedicated to deliver best in-class speech analytics capabilities to contact centres across the world has gone from strength to strength with Castel’s agents, using Intelligent Voice’s powerful GPU powered software, taking over 240000 monitored calls each day on premise. It made sense to enable the same speed and accuracy running in the cloud.
Nigel Cannings, CTO Intelligent Voice

Cloud services, being well known as an inexpensive alternative to on-premise, opens the opportunities and capabilities to additional customers that otherwise may not have the infrastructure to run call monitoring on premise.

About Intelligent Voice®
Intelligent Voice Limited is based in London, San Francisco and New York. The company has over 25 years’ experience of delivering mission critical systems in the financial services industry, including to several of the world’s top 20 insurers and banks. Through innovations such as the SmartTranscript® and GPU-accelerated speech recognition, Intelligent Voice allow companies to understand their businesses better, with a key focus on unlocking the value in telephone and meeting room audio. For further information about Intelligent Voice, please visit www.intelligentvoice.com

About Castel:
Founded in 1982, Castel designs call center software, services and solutions engineered for businesses. Castel listens, learns, plans and partners with companies to define and realize their future. For more information, news and perspectives from Castel, please visit Castel Newsroom at http://www.castel.com/news/.