Specifications

PF1138-02

Text to speech middleware

● Text to speech middleware for the S1C33 Family

● Generates more natural speech by adjusting parameters

for each phoneme

● Comes with VSX2, a high sampling rate version of VSX

■ FEATURES

•The text-to-speech middleware for the S1C33 Family, this generates speech from Japanese text using pho-

neme data compressed by VSX (8kHz) or VSX2 (11.025 to 22.05kHz). Available in linkable library form.

• Speech can also be generated from your original registered words, with support for other languages aside

from Japanese.

•A VSX2 speech compression/expansion technology supporting sampling rates of 11.025kHz, 16kHz, and

22.05kHz has just been introduced.

• The sound volume, pitch, length, and silent length between phonemes can be individually adjusted for each

phoneme, for a more natural speech quality closer to actual speech.

•Running on a PC, the S5U1C330T1S tools allow you to adjust output parameters as well as evaluate sound

quality before final production.

• The standard Japanese language phoneme data can be stored in approximately 100KB of memory (for a

sampling frequency of 16kHz).

• The library program is about 15KB in size.

•Ideally suited for vending machines, PDA, electronic toys, and electronic stationery.

■ HARDWARE CONFIGURATION

S1C33xxx

S1C33000

Core CPU

16-bit timer

Built-in RAM

External RAM External ROM

Bus interface

D/A converter

PWM

Speaker

POWER

AMP

■ SOFTWARE CONFIGURATION

● S5U1C330T1S Library

Appilcation

S5U1C330T1S library

(VSX2 function in S5U1C330V1S library)

Hardware (16-bit timer etc.)

S1C33 Family S5U1C330T1S Middleware