UCbuyer Logo

Your Source For Enterprise Communications & Collaboration News & Insights

3 Ways Business Create Value With Speech-to-Text Technology

Posted by Kevin Gulley

Nov 4, 2014

In today's Unified Communications world, there is more voice-based communications and collaboration happening then ever.  Even though we all text like crazy, phone calls, voicemails, face-to-face meetings, web conferences, video calls, video conferencing and webinars are part of our everyday business landscape and aren't going anywhere.  Voicemail_transcription_mobile_phoneVast amounts of information is conveyed during these conversations, but a surprisingly small amount of this is captured and capitalized on.  One reason is that the spoken-word is not really usable data until you turn it into the written word.  This happens every day with note taking, but we are thinking bigger.  We wanted to learn how businesses are currently taking advantage of speech-to-text technology to turn the spoken word into the written word and how that creates business value.

To find out more about what's going on in this space (which has been an area of interest around here since we worked with the world's largest closed captioning and legal transcription company way back in the late 90's), we connected with Roger Northrop, the Chief Technology Officer at Mutare Software and the veteran of dozens of speech-to-text implementation and integration projects.  "Speech-to-text technology has come a long way in recent years, and businesses are finding many ways to create efficiencies, lower costs and improve service with these improved transcription engines," explained Northrop.  “While no engine will provide 100% accuracy, they have become much more specialized and can deliver great results for a wide range of common and custom applications, especially when the output is integrated with additional databases.”

Voicemail Transcription Leading the Charge

While people are enamored with "Hey Google" and I personally love the voice transcription on my iPhone for writing long texts and emails (my fingers are too fat to use that tiny keyboard), when it comes to businesses and speech-to-text, voicemail transcription is currently the killer app.  As we explained in a post recently, tapping into digitally recorded voicemails and bouncing them against a transcription platform allows the recipient to receive an email or text message within moments containing a voice-to-text translation of the message. This provides the employee with the gist (and often, if the caller doesn’t mumble, the exact transcription) of the voicemail, allowing them to gauge its importance, determine whether they should listen to the entire message for more nuanced insights and prioritize the response time.  All of which can result in better decision making, happier customers and hopefully more opportunity.

According to Northrop, this is quickly becoming a business must-have.  "Businesses like law firms go crazy for this technology.  They are always in meetings, or in court or depositions where they are not allowed to take or make phone calls, but they can check their texts or emails.  When they receive an important voicemail transcription, they can excuse themselves if necessary to return the call."  Real estate firms are another great example of businesses that benefit from this tech.  “They get a ton of voicemails and when they are notified of the message content they can return the important calls much more quickly.”

Related:  How to Boost Client Retention Using Hosted Voicemail to Email and Speech to Text

Contact centers (that already use IVR systems for certain things) are now setting up voicemail boxes so callers can leave a message and have someone get back to them instead of wait around in a busy queue.  "The great thing about this approach is that you can send the agent a transcription of the voicemail prior to their returning the call so they can be up to speed.  Data can also be extracted from the caller's message and integrated with the contact center software so the agent is presented with the correct customer record, or so that the call can be pre-routed to a specialist based on information in the transcription, if necessary," explained Northrop.  

Speaking of Voice-to-Text Integration

This is where speech-to-text can really make an impact.  By converting speech into data, businesses can capitalize on all kinds of custom integration's that save them time, money and increase productivity.  "Once you have the transcription you can trigger events inside any database, and this is where you can really blow the whole concept wide open," explained Northrop, This  could take the form of a "data fill" application in which someone fills out a form on the phone by answering questions and the answers are transcribed and placed in the application. Roger shared a story about a government client that utilized this approach not only to save their constituents time and streamline an outdated process, but also to leverage this newly available data (in this case, prior to the telephone solution the information used to come handwritten via paper forms) to look for fraudulent claims by bouncing it against several additional data sources.  This system paid for itself in days by immediately beginning to identify millions of dollars of bogus submissions.   

Additional examples of integration include doctors being able to trigger an app and speak a command (which would need to be encrypted for HIPAA purposes) stating "Send me the reports for patient number 569413 and patient 228011", which the electronic medical record (EMR) system would then execute.  Integrations with CRM systems are becoming more common and these allow salespeople to leave a voicemail or trigger an app to speak a meeting wrap-up and have it transcribed and connected to the customer record.  They can also use spoken command and control tech to trigger certain events like setting a follow up reminder or sending out a specific mail merge email.  "Once the spoken word is turned into usable data, the only limit is your imagination," says Northrop.

Are Media Companies Next for Speech-to-Text?

With the accuracy and speed of the speech-to-text engines beginning to approach that of human transcriptionists, don't be surprised to see media companies looking to use this technology for closed captioning on the fly in the next couple of years.  This could be done for television, radio shows, or for recording and transcribing conference sessions at tradeshows.  While they're at it, they should time stamp it and make their entire catalog of spoken-word content searchable...there must be all kinds of additional business opportunities in that.  Talk about big data.

What are you seeing out in the market?  How are businesses you aware of leveraging speech-to-text technologies to create business opportunity.  Let us know in the comment box below.



Topics: Voice, video, collaboration, Contact Center, Business Case, healthcare