TECH: First Single-Chip Text-To-Speech Processor

From: J. R. Molloy (
Date: Sun Nov 04 2001 - 11:43:50 MST

Winbond Launches Industry's First Single-Chip Text-To-Speech Solution
IC Utilizes Actual Speech Samples That Enable More Realistic Speech Synthesis
for Low-Resource Consumer Applications

Winbond Electronics Corporation America (WECA), a leading supplier of
semiconductor solutions, today introduced the WTS701, the industry's first
single chip IC that converts Text-To-Speech (TTS). The IC uses a concatenation
algorithm combined with Winbond's proprietary Multi-Level Storage (MLS)
technology, providing a significant advancement in this technology by enabling
OEMs to provide TTS functionality to applications that utilize minimal
computing and memory resources. This means that TTS is now available without
using microprocessors, PCs or computers. Completely programmable, the chip is
capable of supporting any number of languages. Initially, U.S. English and
Mandarin are available. The WTS701 is currently available for sampling and is
priced at under $10.00 U.S. in volume quantities. Volume quantities are
expected to be available in first quarter of 2002.

The WTS701 accepts ASCII (Unicode for Mandarin language) input via a serial
SPI port and converts this to spoken audio via an analog output or digital
CODEC output. The chip features a small footprint, making it ideal for a wide
variety of communications and consumer applications that can be enhanced with
text-to-speech synthesis. Among many applications, the chip can be used to
read email, SMS messages, instant messages, Web pages, news, weather, sports,
and stock quotes; as well as to deliver e-commerce information or directions
with the aid of a GPS location device or other network. It may also be used in
learning-aids, products for the visually impaired, e-books, toys and other
entertainment devices.

TTS chip solutions on the market today typically employ speech synthesis based
on computer-generated models. The WTS701 creates a more natural human sounding
speech by converting text to speech using recorded human speech samples. This
ensures that the output is a recognizable human voice, unlike
computer-generated synthesis. Processing for the WTS701 is performed by,
'concatenating' pre-stored acoustic elements into words. The chip contains two
memory arrays; one for the acoustic library and one for the processor program,
enabling user customization for special characters such as SMS icons or chat

The WTS701 also offers a number of other features including an A/D converter
for sending audio data over an audio PCM bus, and speaker driver circuitry for
analog outputs. Device characteristics include +2.7 to 3.3 Volt power supply,
low power consumption, 56 pin TSOP package and an industrial temperature range
of minus 40 degrees to 85 degrees Celsius. The WTS701 is platformindependent
and will connect to any baseband DSP or host controller using a simple serial
interface. This enables OEMs to reduce time to market while offering a
low-resource, single-chip solution for embedded applications.

The small size and low power consumption enable the WTS701 to be easily
integrated into such applications as mobile phones, hands-free accessories and
other mobile devices, with minimal effect on power consumption or development
time. Many applications currently on the market today feature very small
display screens that are unsafe for viewing as users often access information
while driving.

--- --- --- --- ---

Useless hypotheses, etc.:
 consciousness, phlogiston, philosophy, vitalism, mind, free will, qualia,
analog computing, cultural relativism, GAC, Cyc, Eliza, cryonics, individual
uniqueness, ego, human values, scientific relinquishment, malevolent AI

We move into a better future in proportion as science displaces superstition.

This archive was generated by hypermail 2b30 : Sat May 11 2002 - 17:44:17 MDT