Should AI be Regulated? — AI Show

Scott: Welcome to the AI show. Today we’re asking the question, should AI be regulated? Susan: This is a big one—The government has figured out that AI exists. Scott: Like, “Whoa. Wait a minute. This is important stuff.” Susan: It’s like the government has realized that maybe the world can change and all sorts of… Read More »

Under the weather — What Does it Mean?

We know this means to be sick or ill, but why? Like so many set phrases in English, to be under the weather is a sailing term. There are several conjectures as how it came about. Here are three: A picture the ship’s log from the U.S.S. Seminole, Thursday, November 07, 1861. Back in the… Read More »

The Trouble with Word Error Rate (WER)

In the spring of 2017, Google announced that their voice recognition WER (word error rate) had fallen to just 4.7%—putting it on the par with human transcriptionists. “Wow!” said the world. Highly accurate speech recognition is changing how we interact with computers, how we advertise and so much more. Too bad that a 4.7% WER… Read More »

How Does Santa Do It? — AI Show

Scott: Welcome to the AI show. Today we’re asking the question. The big question, the really big question, the secret deep down thing that we think we’ve uncovered: What is Santa’s secret. Susan: Yeah how? How does that big jolly red guy do it? Scott: How does he do it? We think we figured it… Read More »

How is Machine Learning, or Deep Learning, Affecting Science? — AI Show

Scott: Welcome to the AI Show. Today we’re asking the question: How is machine learning, or deep learning, affecting science? Susan: Actually I’m asking a question of you! For those that do not know, Scott here has a little bit of a science background. Scott: A little bit. Susan: And a little bit of machine… Read More »

Do Your Call Transcripts Read Like Mad Libs?

When the speech recognition API you use fails to recognize the words that matter to you, it produces absurd sounding and often useless transcripts. Such annoying (though sometimes funny) errors can be avoided by using a custom speech recognition model designed to understand your particular call data. Accurate, custom speech recognition APIs can be the… Read More »

What is the Most Important Channel to Engage Your Customers On?

At the Talkdesk Opentalk 2018 Conference, Deepgram had the opportunity to talk to leaders about customer care. The 2018 Talkdesk Opentalk conference made patent one fact: A.I. will revolutionize customer service. We at Deepgram were curious to get the pulse on what people thought of voice, CX and A.I. We interviewed fellow attendees and speakers… Read More »

What Are the Top Mistakes in Deep Learning? — AI Show

Scott: Welcome to the AI Show. Today we’re asking the question: What are the top mistakes in deep learning? Susan: We’ve got huge ones! We make them all the time. Scott: The mistake is the rule in deep learning. You make nine mistakes and one maybe good move. Susan: How are you gonna find new… Read More »

What is Word Error Rate (WER)?

Word Error Rate (WER) is a common metric used to compare the accuracy of the transcripts produced by speech recognition APIs. Speech recognition APIs are used to surface actionable insights from large volumes of audio data in addition to powering robust IVRs and voice-command-enabled devices such as the Amazon Echo. Product developers and data scientists… Read More »

Why Does Your Speech Recognition Need Context?

Alexa and Siri are great and all, but they often make funny, if not serious mistakes. Why is this? I conducted an experiment to measure how important context is in understanding audio—especially when the audio is noisy. As humans we tend to enter conversations with some knowledge of what the topic is, which allows us… Read More »

What’s the Best Infrastructure for Machine Learning — AI Show

Scott: Welcome to the AI Show. Today we ask the question: What’s the best infrastructure for machine learning? Susan: We have seen a revolution recently and we can point to one of the big enablers, which has been GPUs. We’ve talked about this a whole lot. GPUs have not always been here. Scott: Machine learning… Read More »

Why IoT Means Speech Recognition

People now use their voice to control an increasing number of their devices—from TVs to cars, and everything in between. The technology that makes voice control possible is speech recognition — technology that transforms your verbal commands into executable actions via an API. How do Speech Recognition APIs work? When you talk to your speech… Read More »

How Can Companies Extract More Value Out of Voice?

There is a good reason why voice communication is the “gold standard” for interacting with customers. When we communicate with others, we try to understand what is being said by paying attention to three particular areas: the hands, face and voice. Each of these systems were co-opted by evolution for highly complex, fine-grained communication. In… Read More »

How Do You Use a Neural Network in Your Business?—AI Show

Scott: Welcome to the AI Show. Today we’re asking the question: How do you use a neural network in your business? Susan: Well let’s just talk about what people think of neural networks, simple ones. There’s sort of the classic one, the very first thing you ever build when you’re learning how to deal with… Read More »

How to Become a Customer-Centric Organization

Customer-centric organizations understand that their mission is to give their customers good service. The businesses that succeed, have figured out how to do so while making a profit. If you listen to companies like Zappos, the secret sauce, then, is to listen to the customer. Saying that you listen to your customers is one thing,… Read More »

Finally, a reason to get excited about voice recognition.

Check out Deepgram



Artificial Intelligence

Automatic Speech Recognition

Best Practices

Company News

Customer Story

Deep Learning

Voice Recognition

Word Error Rate


Contact Us

Speech recognition is hard. We'll make it easy.