Industry

Top ASR, NLP, and NLU Tools for Revenue Intelligence Platforms

To build a smart, competitive Revenue Intelligence Platform, you'll need to utilize ASR and NLU/NLP tools like Speech-to-Text and Audio Intelligence APIs.

Top ASR, NLP, and NLU Tools for Revenue Intelligence Platforms

Table of contents

In today’s fast-paced, competitive sales environment, sales, marketing, and customer success teams need immediate and complete access to deal visibility. Teams must also be able to easily collaborate to spot risks and opportunities in potential deals, create accurate forecasts, coach reps to meet their numbers, and automate manual tasks.

Revenue Intelligence Platforms can help streamline this process for modern brands.

To build a smart, competitive Revenue Intelligence Platform, you’ll need to utilize the latest Conversational Intelligence and AI-based analytics technology.

This article looks at how tools like Automatic Speech Recognition (ASR) and Natural Language Understanding and Natural Language Processing (NLU/NLP) tools such as Audio Intelligence APIs–created using cutting-edge AI and Machine Learning (ML) research–can help.

Revenue Intelligence Platforms

Before we dive into ASR and Audio Intelligence, let’s first examine what exactly a Revenue Intelligence Platform is and what makes it so important.

What is a Revenue Intelligence Platform?

A Revenue Intelligence Platform uses Artificial Intelligence (AI) and Machine Learning (ML) technology to accomplish three main goals:

  1. Capture customer interactions in real-time via video calls, phone calls, email, and more, and to tie these interactions into CRM data.
  2. Understand conversations across topics, messaging, sentiments, and other important data.
  3. Offer intelligent insights that help teams make better sense out of a wealth of data.

Why are Revenue Intelligence Platforms Important?

Why not just use a traditional CRM to accomplish the above? Unfortunately, today’s sales teams are confronted with an overwhelming abundance of data that’s often outdated by the time it is viewed (most of the data in a CRM is already five days old). Moreover, because this data isn’t collected in an intelligent way, it typically only represents 1% of customer opinions or actions, so the data isn’t representative of a customer base anyway.

Together, this practice muddies our view into what customers are really thinking and doing, rendering decisions made based on this data educated guesses at best.

Revenue Intelligence Platforms seek to solve this challenge by using AI and ML to find patterns and intelligent insights within this data.

According to Gong, there are five main principles of Revenue Intelligence:

  1. To replace manual data collection with automation, AI, and ML.
  2. To replace limited views of customer interactions with the whole picture.
  3. To acquire real-time data for intelligent decision making.
  4. To make decisions based on true customer opinions and actions that are supported by data.
  5. To connect an entire company and its decision making through this data.

While that may sound like a lot to accomplish, today’s advances in AI, ML, and Deep Learning (DL) research have made Revenue Intelligence more attainable.

Let’s look at how two of these AI-backed technologies–ASR and NLP/NLU–help facilitate this process.

Speech-to-Text and Audio Intelligence for Revenue Intelligence

Highly accurate ASR platforms such as Speech-to-Text APIs automatically transcribe video and audio streams, like sales calls, into a readable transcription text. In the past, however, this meant a large block of transcribed text that was free of punctuation and casing, paragraph structure, and speaker labels, making transcriptions cumbersome to read through.

Thankfully, the best Speech-to-Text APIs today can automatically alter a transcription text to include the elements listed above, making a text much easier to digest. AssemblyAI’s Automatic Casing and Punctuation Models, for example, are trained on text with billions of words for exemplary transcription accuracy and utility.

Some Speech-to-Text APIs also go beyond this basic transcription, offering advanced NLU/NLP tools referred to as Audio Intelligence APIs.

Built using the latest ML, DL, and NLP research, Audio Intelligence APIs let users quickly build high ROI features and applications on top of their audio data–helping companies move past line level transcription. This could include detecting important entities in a text, identifying sentiments spoken by speakers, sorting a text automatically into chapters, and more.

Let’s now look more closely at the biggest impacts this ASR and Audio Intelligence technology can have on Revenue Intelligence Platforms and software:

1. Automatically Identify Important Sections of Calls

Revenue Intelligence Platforms must be able to automatically identify important sections of calls–whether it be places where important questions are asked, objections are raised, or where changes in tone occur. These platforms also need to be able to intelligently analyze conversations to determine buying patterns and identify potential churn risks.

Three main Audio Intelligence APIs can help support these needs: Auto Chapters, Topic Detection, and Sentiment Analysis.

Auto Chapters, or Summarization, provides a “summary over time” for transcription texts. It works by (A) segmenting the text into logical chapters, or where the conversational topic changes, and then by (B) automatically generating a summary for each of these chapters.

Auto Chapters is a great tool for making calls easier to skim and/or navigate, especially when performing QA.

Topic Detection identifies common topics in a transcription text, helping users understand context and identify buying patterns.

Topic Detection can help determine if certain topics do (or not) lead to more sales, cause questions or objections, and more.

Sentiment Analysis detects positive, negative, or neutral sentiments in a transcription text, by speech segment and/or speaker. For example, the speech segment:

I love traveling to New York City. 

would be labeled as a POSITIVE sentiment.

Sentiment Analysis APIs can help identify changes in tone in a transcription–where did a call change from positive to negative? Are more positive sentiments associated with certain topics? Then, this data can be used to determine patterns and identify corresponding actions.

2. Increase Efficiency on Digital Selling Tasks and Generate Deal-specific Insights

Revenue Intelligence Platforms also need to efficiently synthesize data to generate actionable, deal-specific insights for sales teams. AI-powered tools like ASR and Audio Intelligence can assist here as well by generating real-time data for strategic analytics, which can be leveraged to better identify and score winnable deals.

Three Audio Intelligence APIs help automate and generate these insights: Entity Detection, Auto Chapters, and Sentiment Analysis.

Entity Detection (A) identifies and (B) classifies specified information in a transcription text. For example, lawyer is an entity that is classified as occupation. Common entities that can be detected include personal and organization names, locations, addresses, phone numbers, languages, dates, nationalities, and more.

Entity Detection can help automatically sort through a transcription text to uncover gaps in the sales process, identify trends, and turn this data into actionable insights.

Similar to what is described above, Auto Chapters can also help facilitate a smarter sales process by helping teams identify trends in conversations. Sentiment Analysis can provide an overview of attitudes towards products, events, or even specific sales representatives.

3. Optimize Coaching with Relevant Data Insights to Keep Customers Engaged

Finally, Revenue Intelligence Platforms should also help optimize a sales team's performance and boost customer engagement and experience.

All four of the Audio Intelligence APIs discussed previously–Auto Chapters, Topic Detection, Sentiment Analysis, and Entity Detection–can be used together to build products that help sales teams better understand the true impact of every sales conversation.

Do customers respond better to certain topics or entities? Do certain agents have more success employing positive or neutral sentiments? Audio Intelligence unlocks this insight in close-to-real time. Then, this data can be analyzed and aggregated to inform whether to replicate and scale the successful behaviors or refocus and retrain on the less successful ones.

AI-based Revenue Intelligence Platforms

Sales, marketing, and customer success teams must be equipped with powerful tools to win deals in a saturated market.

By incorporating AI and Machine Learning-backed tools like ASR and Audio Intelligence APIs, you can create a smarter Revenue Intelligence Platform that automates conversation transcription and provides strategic, high utility analytics that helps teams meet–and exceed–sales goals.