Transcription: The Hidden Weapon To Accelerate Transformational Change


Share on LinkedIn

Artificial intelligence is now making real-time transcription and speech analytics investments in contact centers meaningful because it allows customer experience (CX) operations to make radical improvements.

Imagine capturing 100% of every customer interaction, whether voice or digital. Imagine having objective, real-time insight into drivers behind your contact center metrics. Thanks to recent development of efficient training mechanisms and the availability of more computing power, deep neural networks have enabled Automatic Speech Recognition (ASR) systems to perform astoundingly well.

I spent virtually all of my professional career working with CX and IT leaders in Fortune 500 companies that invested millions of dollars annually in speech analytics, but the benefits maxed out years ago for delivering quality management improvements. A consistent theme when re-evaluating the speech analytics toolset was: “There are no new insights into what customers are contacting us about. We get reports – but there’s nothing really new – so nothing changes.” Leveraging AI across 100% of the CX data set to identify previously unrecognized patterns is revealing opportunities previously hidden just below the surface.

ASR advancements were evident at Interspeech, the largest technical conference on speech processing and application held last month. For the last five years, LibriSpeech, a speech resource of 1,000 hours of transcribed audiobooks, has been the most used benchmark dataset for ASR research. Facebook, ASAPP and Google have all broken the 98% speech accuracy barrier in 2020 and the innovation race will continue.

Automatic Speech Recognition (ASR)

Multi-stream Convolutional Neural Networks (CNN) where audio is handled for better responses to noisy audio – is one of the main contributing factors to the research outcomes. These improvements allow ASR systems to take advantage of models that offer reliable transcriptions for noisy environments like agent-customer conversations.

Ultimately these innovations benefit consumers, as 100% real-time transcriptions feed directly into benefits for companies in numerous industries—whether it’s augmenting and amplifying customer service agents’ skills, or to transcribe videos or talk-to-text. Contact centers that support both voice and digital channels will certainly benefit from ASR as a backbone for transcription, voice of the customer (VoC), customer sentiment, and coaching.

AI-driven analytics, and the ability to quickly act on insights can be a hidden weapon to accelerate transformational change within the organization.

Transcription. 100%. Real-Time

Most companies record, batch and store between 10-20% of calls for later analysis and transcription. These historical, after-the-fact, records are not an accurate representation of everything happening across the totality of CX operations and result in business decisions being made based on a narrow dataset. With advanced ASR systems, 100 percent real-time transcriptions can be achieved, which is a game changer for consumer companies aiming to dramatically enhance customer loyalty, retention, and revenues by way of highly modernized and intelligent CX operations.

Pairing transcription with real-time AI-driven analysis also makes it possible to support and prompt customer service agents with the best responses and actions to take with customers based on machine-learning. It allows the Voice of the Customer (VoC) insight to be captured from every call and enables automation of thousands of micro-processes and routine tasks, like call summaries. One major telecom company in North America is already leveraging AI to automate the dispositioning of notes at the end of each agent call to realize a 65% reduction in handling time for that specific task. Leveraging these capabilities to enhance customer and agent experiences in both sales and service environments will enable companies to break through the ceiling of benefits experienced with legacy technologies and the intelligent insights will fuel ongoing optimization of operations well into the future.

Customer Sentiment Analysis and Prediction

Customer Satisfaction (CSAT) and Net Promoter Scores (NPS) surveys capture feedback response rates from 5%-15% of customers in most CX operations. With innovative machine learning algorithms leveraging transcriptions and speech analytics of every interaction, CX leaders can now predict both the sentiment and satisfaction of customers – enabling agents to increasingly get it right the first time rather than hoping you can repair the brand after a poor interaction.

Data can be unearthed to discern trending intents of customers’ calls made to the company in real-time. If you don’t know the specific drivers behind the interaction metrics within your company, it’s nearly impossible to make dramatic advancement in your CX programs.

By automatically categorizing each purpose of customer inquiries, CX leaders can obtain a ground-to-cloud-level view of what’s happening with customers. These AI advancements let CX leaders analyze the most common topics, trends, and performance from tens of millions of customer interactions that occur every year. From call duration, wait times, handle times etc., and the key drivers behind those metrics: What were the reasons for unexpectedly long handle times? Were agents clicking around the knowledge management database trying to find answers? Or how about unbiased opinions on cancel rates? Were they due to a product flaw or issues with customer service? Could better save approaches have been used? Or what may have caused the customer sentiment to shift during the interaction?

This informed understanding of why customers are calling delivers insights and trends that departments across the entire enterprise can use to make decisions—all based upon data. As American computer scientist, Rear Admiral Dr. Grace Murray Hopper said: “One accurate measurement is worth a thousand expert opinions.”

When businesses can adjust and respond to consumers quickly—even applying anomaly detection to identify real-time crisis issues and address them before they become disastrous —customer revenues, loyalty and satisfaction all rise.

The radical improvement of ASR accuracy will allow enterprises to reduce their overall CX spending by 30%, which can translate into hundreds of millions of dollars annually and accelerate the transformational change that CX leaders are seeking. CX has rapidly become a battleground in a highly competitive marketplace – highly accurate, real-time transcription holds the potential to make CX your competitive advantage.

Chris Arnold
Chris is Vice President of contact center strategy at artificial intelligence firm ASAPP that is automating and augmenting human workflows. Chris was formerly Verizon’s Global Director responsible for modernizing and orchestrating technology for 50,000 contact center agents operating in voice, chat, messaging, and social media engagement functions.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here