Speech-to-Text Selection – David Borish, PRIMO AI – Voice Tech Podcast ep.038

Episode description

David Borish is Chief Creative at PRIMO AI, a New York startup that recommends the highest performing Speech-to-text (STT) and Natural Language Understanding (NLU) services for a particular dataset and geographical region. We discover what the biggest problem with speech to text systems is today, and why trying to solve it by hiring data scientists can be prohibitively expensive.

We also discuss the advantages of acquiring a technology patent, why David chose to recently enter the voice space, and the approach he takes when selecting his next entrepreneurial challenge. David is a seasoned startup veteran who believes passionately in the future of voice, and our conversation contains many valuable lessons to take away.

Highlights from the show

What’s the biggest problem with speech to text systems today? They don’t understand words spoken in dialects very well. They can fail to disambiguate words/phrases that have different meanings across dialects.
What is a dialect? A specifc form of speech within a language, created with different cultures combine. There are 6000+ dialects around the world.
Why are cloud-based Speech-to-text services quite cheap to purchase? They are a data play, as you’re helping them to build their service by supplying your audio data.
Why don’t companies train their own Speech-to-Text systems? Even companies with limited speech domains like banks must train a model with huge amounts of data to capture all the variety in speech across dialects.
How can PrimoAI help? It scores different STT and NLU providers peformance on your own speech audio data, helping you find the best overall service, and the best for each segment of your customer base.
Why not do it yourself? The advantage of PrimoAI is you don’t need to hire an expensive data scientist to perform the performance comparisons on each service.
What’s the advantage of acquiring a technology patent? It provides some strategic defense to dissuade new competitors from entering your market, and it also adds value to a company at exit.
What partnerships does Primo AI have? Currently working with the Google ML team, identifying customers who are unhappy with the performance of Google products.
Why did David choose to enter the voice space? He passionately believes voice is the future of computing. So he looked at where the problems currently lie, and found STT systems were underperforming for his family and people in industry.
What are the things to look for in a problem that’s worth building a business in? Find areas that you don’t have experience in, so you can have fresh ideas and think creatively. You must also start with a natural curiosity, to you the drive to dig into the subject and go the distance.
How to test whether your idea is promising? Talk to many smart people and experts in the field, to get feedback on your idea. Always room for new competitors in the market; you don’t need to be the first to succeed.