AI and Machine Learning: The Power Behind Today’s Best Customer-Research Platforms

September 19, 2022

The customer-feedback industry is exploding, with its market size expected to reach $3,292.8 million USD by 2027. Many companies that want to capitalize on this growth are turning to customer-research platforms. These platforms can compile huge datasets of qualitative and quantitative video and voice feedback to generate rich analysis and targeted insights for these companies.

For example, a customer-research platform might compare product messaging to customer descriptions, offer an analysis of customer sentiment to inform a company’s public-facing decisions or leadership changes, or utilize thematic clustering to analyze themes across qualitative survey responses to inform action items or even KPIs (Key Performance Indicators).

The foundation of these customer-research platforms is artificial intelligence (AI) and machine-learning technology, including automatic speech recognition (ASR) and natural-language processing and natural-language understanding (NLP/NLU) applications.

Champion Advertisement
Continue Reading…

The accuracy and accessibility of automatic speech-recognition applications such as speech-to-text APIs (Application Programming Interfaces) has dramatically increased in recent years, automating the generation of highly useful, digestible transcripts for companies and platforms alike.

Then, on top of this transcription data, platforms are offering natural-language processing and understanding applications, which some refer to as audio-intelligence tools, to identify trends across feedback data that human analysis might not have surfaced.

This article examines the three main ways in which ASR, NLP, and NLU—which are backed up by cutting-edge AI and machine-learning research—are transforming today’s best customer-research platforms, as follows:

  1. By facilitating a more efficient review process
  2. By surfacing key highlights, themes, and trends
  3. By creating smart tags to categorize and search for information

Facilitating a More Efficient Review Process

The best speech-to-text APIs today integrate with customer-research platforms to transcribe both asynchronous and live voice or video customer feedback, with nearly the same accuracy as a human transcriber as measured by word error rate.

Word error rate (WER) is the de facto standard of measurement for accuracy in speech recognition. Essentially, we can calculate WER by measuring the number of errors in the text of a transcription versus that of a human transcriber. While WER is a great starting point for comparing accuracy across transcriptions, keep in mind that it doesn’t consider other factors such as transcript readability, context, the use of text versus numbers—for example, seven versus 7; capitalization, and paragraph structure.

Thankfully, today’s speech-to-text APIs don’t just supply a wall of text. Imagine reading a novel without quotation marks, he or she said markers, paragraphs, or punctuation. Reading it would take an enormous amount of effort, right? Unfortunately, that is how speech-to-text APIs used to output transcripts.

But speech transcription has come a long way since its inception. In addition to its significantly higher accuracy, today’s speech-to-text APIs produce formatted documents that include automated punctuation and case, paragraph structure, and speaker labels, if applicable. This automated formatting greatly increases transcripts’ readability, as well as their utility.

In addition, some speech-to-text APIs can redact personally identifiable information (PII) from text automatically, in the event that companies are obligated contractually or legally to redact personal or sensitive information. The redaction of PII could include phone numbers, addresses, social-security numbers, and credit-card numbers, replacing each number in a sequence with a # symbol.

Through automated transcription, readability features, and PII redaction, customer-research platforms can already facilitate a much simpler review process for the companies they serve.

Surfacing Key Highlights, Themes, and Trends

Once customer-research platforms have accurately transcribed video and voice feedback, they can apply NLP/NLU tools to identify important highlights, themes, and trends across thousands of customer responses. Let’s look at some of these NLP/NLU tools.

Topic Detection

Topic detection APIs identify and classify topics, according to the IAB Taxonomy, a four-level classification system, listing about 700 commonly identifiable topics. Topic detection APIs help customer-research platforms to identify important, commonly recurring topics across transcripts.

Entity Detection

Similar to topic detection, entity detection APIs identify and classify the entities in a transcription text. For example, New York City is an entity that would be classified as a location. Entity detection APIs help customer-research platforms identify important, commonly recurring entities across transcripts.

Sentiment Analysis

Sentiment analysis APIs label the speech segments in a transcription text as positive, negative, or neutral. For example, the speech segment I enjoy going to the movies would be labeled as positive. Sentiment analysis helps customer-research platforms to analyze the feedback they’ve collected to identify commonly recurring sentiments about services and products.

Text Summarization

Text summarization APIs provide informative summaries of lengthy transcriptions, making them more digestible and useful. Typically, text summarization provides a short headline for each section in which the conversation naturally changes subject or topic, along with a multi-sentence summary of the discussion under each headline.

Together, these intelligent NLP/NLU APIs capitalize on the latest AI and machine-learning research to surface true insights into respondents’ attitudes, behaviors, and actions.

Creating Smart Tags to Categorize and Search

Finally, customer-research platforms can use the APIs I’ve described to categorize the research data and survey responses. These categories can then feed into a smart-tag system, enabling companies and users to search the tags to find the data and responses they need—similar to the use of hashtags on Twitter.

Platform users can then aggregate this tag data by adding sophisticated analysis tags and even generate reports based on these specified tags. Therefore, smart tags can be an extremely sophisticated analysis tool.

The Symbiotic Relationship of AI and Research

AI and machine learning have made massive strides in the past few years, thanks to cutting-edge new models and research methods. These advances have also had significant downstream effects on research areas such as automatic speech recognition, natural-language processing, and natural-language understanding.

Industries are now taking notice, incorporating some of these intelligent applications into their platforms and software—thus, creating some pretty powerful tools for their users.

Customer feedback and research is one of these industries. By incorporating ASR and NLP/NLU applications such as accurate speech transcription, topic detection, entity detection, sentiment analysis, text summarization, and smart tags, customer-research platforms can create competitive offerings that directly impact customer satisfaction, product success, and companies’ earnings. 

Content Marketing Specialist at AssemblyAI

Albuquerque, New Mexico, USA

Kelsey FosterKelsey is an experienced technical writer. An academic at heart, she continuously strives to learn more about the world around her and help others make sense of complex topics. She currently researches and writes across software development, APIs (Application Programming Interfaces), automatic speech recognition, and more.  Read More

Other Articles on Artificial Intelligence Design

New on UXmatters