IEEE Circuits and Systems Magazine - Q3 2023 - 4

4) Innovation: answering that question.
5) Business: making money on the answer.
I learned to focus on the third statement.
Question 2: What inspired you to start working on digital
signal processing (DSP) at Texas Instruments in the
1970s, particularly when the technology was still primarily
limited to academic research? Could you discuss
the challenges you faced during the development of TI's
first LCD calculator and Speak & Spell, and how you overcame
these obstacles? What strategies did you employ
to ensure the success of these innovative technologies?
Gene Frantz: I joined TI in their calculator division as
a product engineer. It was my job to keep a calculator
production line running with good yields. I had a friend,
Larry Brantingham, who got to play with new, crazy
ideas of which our senior management had thought up. I
decided I would " volunteer " some of my time to help him
turn the crazy ideas into reality. At about the same time
I was moved out of product engineering to a new team
looking at educational products rather than calculators,
reporting to a man by the name of Paul Breedlove. As
we were taking a math learning aid by the name of the
" li'l Professor™ " into production, Paul had a crazy notion
that we might be able to create a similar product to
teach children to spell just as the Li'l Professor taught
them to add, subtract, multiply, and divide. A new researcher,
Richard Wiggins, hired into TI who knew about
a new speech technology called Linear Predictive Coding
(LPC). The four of us joined forces to demonstrate
the idea of using synthetic speech to present the words
to the child. (Note: I wrote a book on the development
of the Speak & Spell™. You can find it online: https://archive.org/details/cnx-org-col11501.)
As Larry and I had
just completed TI's first LCD calculator, we were ready
to take on another hard (if not impossible) task.
So, our initial team was made up of a speech researcher
(Richard), IC architect (Larry), systems
designer (me), and our boss with an impossible idea
(Paul). As I tell the story, the task was deemed to be so
impossible we weren't given any research and development
funding and our senior management distanced
themselves from the obvious failure about to happen.
With Paul's connections within TI, he finally received
$25,000 to demonstrate the possibility of a spelling
learning aid. A year and a half later our expanding
team designed four new integrated circuits, the electronic
system, the software that made it work and introduce
the Speak & Spell™ to the world at the Summer
Consumer Electronics show in Chicago in 1978. My
cynical explanation of the first indication that we were
about to be successful was the new claim of ownership
4
IEEE CIRCUITS AND SYSTEMS MAGAZINE
from our management on its success. It seems that it
was the best way for our management to support the
program-leave the team alone and let them sink or
swim without managements " help. "
As the Speak & Spell was my second time facing
the impossible (the LCD calculator being the first), I
began searching for the next impossible opportunity.
By the way, I'm still looking for that next impossible
opportunity. Many years later I was asked by a magazine
editor " what was my favorite product development
of which I was part. " My answer was, and still is,
" the next one. "
Question 3: Could you share more about Texas Instruments'
marketing approach in promoting DSP technology,
particularly in terms of creating a developer environment
and providing strong customer support? How
did these strategies align with the diverse range of applications
for DSP technology?
Gene Frantz: After success with the Speak & Spell, TI's
first general purpose Digital Signal Processor device developed
was the TMS32010. We had designed it to take
on the world of voice and speech. What we quickly discovered
was our customers were using the DSP device
for modems, 3D imaging, and Hard Disk Drives. At that
moment we decided the best solution was to change our
marketing focus from voice and speech to Modems, 3D
imaging and Hard Disk Drives. Our next DSP offering
made Modems, 3D imaging and Hard Disk Drives easy
to design. Surprisingly, our customers " blind-sided " us
again using this new generation of DSPs to solve a new
set of problems. We quickly found that our best marketing
approach was to keep up with the innovations our
customers were making using our DSPs.
For many years I was part of a three-person team
whose task was searching for new uses and new users
of DSP. Virtually all these customers were small startup
companies or skunk works teams inside of large companies.
In a typical customer visit, we would ask our field
salespeople to help us arrange the meetings. Typically,
they were reluctant to help as it would be a distraction
from their existing customers. But they would agree to
pick us up at the airport, take us to the customer(s) and
then return us to the airport. We learned when an interesting
market was emerging based on the sales-person's
weekly report describing how they had uncovered an
exciting new customer.
This exercise of searching for new uses and new users
gave us a competitive advantage in many cases, as
we were finding new market segments forming before
the companies in the market segment knew their market
segment existed. I must note that this team of three was
THIRD QUARTER 2023

http://www.archive.org/details/cnx-org-col11501 http://www.archive.org/details/cnx-org-col11501

IEEE Circuits and Systems Magazine - Q3 2023

Table of Contents for the Digital Edition of IEEE Circuits and Systems Magazine - Q3 2023