5.6 Second LOG-PCM Speech
Features
•
Operating voltage: 2.4V ~5.0V
•
Directly drives an external transistor
•
Low standby c urr en t ( 1µ A ty p. for VD D= 3 V)
•
Minimal external components
•
380 words table ROM for key functions
•
Programmable silence length and end-pulse
width (minimal end-pulse width is 330
6kHz sampling rate)
•
5.6-second voice capacity (based o n a 6kHz
sampling rate)
•
Controllable volume
•
FLAG1 options:
–
End-puls e output
–
3HzB flash
–
6HzB flash
–
Voice output indication
–
Busy output
•
FLAG2 options:
–
3Hz flash
–
6Hz flash
–
Busy output
µs at a
HT813D0
•
8 keys
•
Key options:
–
Stop key: KEY8
–
Random (only for KEY1)
–
Sequential (only for KEY1)
–
Repeat (for all KEYs)
–
Key debounce time (for all KEYs): 700µs,
22ms, 45ms, 180ms (based on a 6kHz sampling rate)
–
One shot (for al l KEYs )
–
Level-trigger
–
Pull-high resistance (for all KEYs)
•
Section options:
–
Retriggerable
–
Non-retriggerable
•
Dice form or 16-pin DIP/SOP
Applic a tions
•
Toys
•
Alarm clocks
•
Public address system
General Description
The HT813D0 is a single-chip LOG-P CM voice
synthesizer LSI wi th 5.6-second voice capacity
at 6kHz sampling rate. The chip when triggered
drives a speaker through an external transistor
with a current switch D/A converter output.
Negligible current will be consumed in the
standby state.
The HT813D0 provides 8 key inputs and 2 programmable FLAG outputs. With 2.4V~5.0V
power supply, a complete synthesized voice
playback system can be ea sily built with very
few external components.
•
Alert & warning system
•
Sound effect generators
The custome r’s voice s ources are reco rded section by section into an internal mask ROM. The
instructions of section playback arrangement of
each key are stored in the table ROM. The key
features are also programmable. With such a
flexible structure, the HT813D0 is excellent for
versatile voice applications.
1 5th May ’98
Pin Assignment Block Diagram
HT813D0
Pad Coordinates
Pad
No.
1 –887.00 702.20 9 903.50 365.80
2 –887.00 524.50 10 886.90 568.00
3 –1007.00 –804.00 11 886.90 743.00
4 –231.30 –754.00 12 626.90 754.00
5 59.80 –754.00 13 76.80 754.00
6 523.50 –754.00 14 –253.70 754.00
7 831.30 –754.00 15 –627.20 754.00
8 1006.90 –804.00
Chip size: 2290
* The IC substrate should be connected to VSS in the PCB layout artwork.
× 1940 (µm)
2
XY
Pin Description
Pin No. Pin Name I/O
1 OSC1 I — Oscillator input pin
2 FLAG1 O
3 VSS I — Negative power supply (GND)
4 FLAG2 O
Internal
Connection
NMOS
Open Drain
NMOS
Open Drain
Description
3HzB/6HzB flash output, busy output, end-pulse or
voice output indication (by mask option). Open
drain, active low output
3Hz/6Hz flash output or busy output (by mask
option). Open drain, active low output
Pad
No.
Unit: µm
XY
2 5th May ’98
HT813D0
Pin No. Pin Name I/O
Internal
Connection
Description
Trigger key, active low. Key features such as
5~12 KEY1~ KEY8 I Pull-High
debounce time, pull-high resistance and repeat can
be selected by mask option.
13 NC — — No connection
14 VDD I — Positive power supply
15 AUD O
PMOS
Open Drain
Voice output for driving an external transistor
16 OSC2 O — Oscillator output pin
Absolu te Maxim um Ratings *
Supply Voltage ............................... –0.3V to 6V Storage Temperature ............... –50°C to 125°C
Input Voltage ............... V
–0.3V to VDD+0.3V Operating Temperature ............. –20°C to 70°C
SS
*Note: These are stress ra tings on ly. Stresses exceeding the range spe cified un der “Abso lute Maxi-
mum Ratings” ma y cause substantial damage to the device. Functional operation of this
device at other conditions beyond those listed in the specification is not implied and prolonged
exposure to extreme condition s may affect device reliability.
Electrical Cha racteristics (Ta=25°C)
Symbol Parameter
V
I
DD
I
STB
I
O
I
OL
V
V
f
OSC
Operating Voltage — — 2.4 — 5.0 V
DD
Operating Current 3V
Standby Current 3V — — 1 3 µA
Max. AUD Output Current 3V VOH=0.6V –1.5 –2 — mA
FLAG Sink Current 3V VOL=0.3V 1.5 3.0 — mA
“H” Input Voltage — — 0.8V
IH
“L” Input Voltage — — 0 — 0.2V
IL
Oscillati ng Frequency 3V R
Test Conditions
V
DD
Conditions
No load,
f
=96kHz
OSC
=530kΩ 76 96 116 kHz
OSC
Min. Typ. Max. Unit
DD
µA
V
—200400
—VDDV
DD
3 5th May ’98
Functional Description
The HT813D0 is a mask ROM type voice synthesizer with 5 .6-second voice capacity. A group of
pre-recorded voice sections is played upon receipt
of key trigger input signals. Two FLAG signals are
output while playing voices.
The 5.6-second voice capacity can be divided into
sections of arbitrary length. Notice that the silence length and end-pulse width are not included in the memory.
By using HOLTEK’s programming tools, the
contents and arrange ment of sections, as well
as key features and FLAG output are all programmable before device fabrication.
The IC provides 8 key inputs (KEY1~KEY8). Of
the 8 keys, KEY1 can b e optioned as a direct,
sequential or rando m trigger key. KEY8 can be
selected as a stop or a direct key. The remaining
6 keys (KEY2~KEY7), are u sed as direct keys
exclusively.
Play function block diagram
Syste m os c ill ator
The HT813D0 has a built-in RC oscillator which
requires only one external res istor for normal applications. The oscillator frequency is typically
96kHz for an external resistor of 530k
quired oscillator frequency m ay vary with different sampling rates in the process of voice
programming . As a res ult , the v alu e of th e osci llator resistor may be different with respect to different items.
Ω. The re-
HT813D0
The oscillator is turned on when triggered by a
key input. After playing, it is immediately turned
off. Then the chip goes into the standby state.
Voice ROM
The voice ROM is originally designed to continuously record the 5.6-second voice data at about
6kHz sampling rate. A higher sampling rate will
generate voices of better playback quality , but will
shorten the total recording time. On the other
hand, a lower sampling rate will result in longer
recording time but sacrifice voice quality.
The playback time can be significantly ex tended
by making use of coding efficiency, silence playing,
section repeating, section cascade, et c.
Section
Section is the basic element constituting the
contents of the voice ROM. During programming, the custome r’s voice sources can be divided into as many sections as required. A
section can be composed of a voice or an interval
of silence. However, the silent length is not
counted in voice ROM. The total number of
sections included should be less than 380 due to
the space limitation of the function table ROM.
The total length of the included sections is limited by the voice ROM.
A section, when triggered by a key input, can be
played once, repeatedly or cascaded with other
sections, depending on the key function table
instructions.
4 5th May ’98
HT813D0
The following are examples of section division:
In addition, a section can be set as retriggerable
or non-retriggerable depending on code option.
•
Retriggerable
When the currently playing section is set as
retriggerable, it will stop immediately upon
receipt of other key inputs.
•
Non-retriggerable
When the currently playing section is selected
as non-retriggerable, it will go on playing till the
whole section is completed, whether or not there
is a key input in the process of playing.
For a key group, some section(s) can be set as
retriggerable and some as non-retriggerable.
When a retriggerable section of a key group is
playing, any key can be triggered to interrupt its
playing. On the other hand, if it is a non-retriggerable section playing, any key interrupt is
invalid.
Group
The HT813D0 plays groups according to the key
input. A group can be made up of on e or more
sections. When a key is triggered, the corresponding group is played. For example, triggering KEY2 plays group 2, and so on. The same
section is allowed to appear in different groups.
However, KEY1 can be made up of multiple
groups when it is optio ned as a sequential or
random key. Otherwise, each key is comp osed
by one group only.
function table conta ins group information and
the playing order of section s in the grou ps. No tice that the tota l amount of sectio ns included
in the groups should be less than 380–the space
limitation of the function table ROM.
•
KEY1 as a direct key
Each key is mapped to a group in the function
table. If a key is not used, the group mapped
to that key is a piece o f sil ence. The following
is an example of the function table:
Group 1 sec.1 + sec.2 + sec.3 + sec.5
Group 2 sec.3
Group 3 sec.2 + sec.2 + sec.3 + sec.4
Group 4 sec.5 + sec.3
As illustrated in the above table, voice R OM
is composed of 5 secti ons, and 11 sections for
the function table. If KEY1 is momentarily
triggered, section 1, section 2, sectio n 3 and
section 5 are played in sequence and then
stopped. Triggering KEY2 plays section 3, and
so on.
•
KEY1 as a sequen tial or random ke y
When KEY1 is optioned as a sequential or
random key, KEY1 can include multiple
groups (sub-groups) in the function table.
However , the remaining 7 keys ( KEY2~KEY8)
are used as direct keys exclusively and comprise only one group in the function table.
An example is show n below:
Group 1-1 sec.4 + sec.2
Group 1-2 sec.1 + sec.3
:
:
:
:
Group 1-N sec.2 + sec.3
Group 2 sec.2 + sec.3
Group 3 sec.3 + sec.5
Group 4 sec.1 + sec.5 + sec.2
Key function table
The sections in voice ROM are played according
to the instructions of the key function table. The
As indicated i n th e abo v e ta bl e, KEY 1 can be
made up of sub-groups. The corresponding
sub-groups is played in sequence each time
KEY1 is triggered.
5 5th May ’98
HT813D0
♦
The playing sequence of sequential KEY1 is:
Group 1-1
→ Group 1-2 → Group 1-3 ..... →
Group 1-N (the last group) → Group 1-1 .....
♦
The playing sequence of random KEY1 is:
Group 1-3
→ Group 1-5 ..... → Group 1-N →
Group 1-3 → Group 1-5 .....
That KEY1 functions as a random key is a
special case of sequential key , which combines
a particular arrangement of sub-group playing sequence.
♦
Reset of KEY1 playing sequence
If a sub-group of KEY1 group is being played
and one of KEY2~KEY8 is triggered, the
playing sub-group will be terminated, and
the newly triggered key group come is played
instead. The first sub-group will start playing by retriggering KEY1. In other words,
the KEY1 playing sequence is reset whenever a key other than KEY1 is triggere d (see
Figure 1).
♦
Sub-group selection
When KEY1 is triggered with pulses, a desired sub-group can be selected by controlling
its corresponding pulse number . However, the
features of KEY1 have to be set in the f ollowing ways:
–
sequential or random
–
retriggerable
–
minimum key debounce time
(
≅700µs, f
OSC
=96kHz)
For instance, if sub-group 1-3 is the previous
playing group, sub-group 1-5 will start playing
after 2 pulses are appli ed to KEY1, an d so on.
To make selection of KEY1 sub-groups more
easily, one of KEY2~KEY8 should be programmed as silence. Then this silence key has
to be triggered to reset KEY1. By so doing, the
playing sub-group of KEY1 is directly specified by the pulse number applied to KEY1 (see
Figure 2).
•
KEY8 as a stop key (by mask option)
When KEY8 functions as a stop key , any voice
output can be stopped by pressing KEY8.
•
KEY1~KEY7 as a repeat key
KEY1~KEY7 all functi on as repeat key if one
of the seven keys is set as a repeat key. In
other words, once the mode of one of
KEY1~KEY7 is determined, the remaining
six keys are set accordingly.
As a repeat ke y, the sections incl uded ca n be
played sequentially and repeatedly till other
key input is triggered. KEY8 has no other
choice but functions as a stop key when
KEY1~KEY7 are set as repeat keys.
Figure 1 Reset of KEY1 playing sequence
Figure 2 KEY1 sub-group selection
6 5th May ’98
HT813D0
Key features
•
Key priority
When two or more keys are triggered simulta-
neously , the output voice is determined by the
key priority as shown below:
KEY1>KEY2>......KEY7>KEY8
•
Key debounce time
There are four kinds of key-in deboun ce time to
be selected by mask option, namely , 700
µs, 22ms,
45ms and 180ms. The k ey de boun ce tim e var i es
with the value of the system frequency.
•
Pull-high resistance
Four kinds of key inp ut pin pu ll -hig h resis-
tance can be selected by mask option,
namely, 20k
Ω, 50kΩ, 100kΩ and 200kΩ.
The resistance may vary with VDD, temperatures and the chip itself due to process
variations.
•
Trigger mode
All of the eight keys can optioned as one-shot
trigger mode or level-trigger mode.
♦
One shot
When one of the eight keys (KEY1~KEY8)
is pressed mom entarily or held down, the
group corresponding to that key will play
once.
FLAG
When voices are playing, both FLAG1 and
FLAG2 pins are activated to output one of the
following signals through code option.
FLAG1 can be o ptione d as on e of the fo llowing
signal outputs:
None, 3HzB flash, Busy, 6HzB flash, Voice indicator , or End-pul se output
FLAG2, on the other hand, can be set as one of
the following signal ou tputs:
None, 3Hz flash, 6Hz flash, or Busy output
•
3Hz/3HzB flash
When voices are playing, FLAG1 as well as
FLAG2 pin outputs a 3 Hz signal to drive a n
LED. The signal is active low, 25% duty. Once
the voice output is terminated, the FLAG1
and FLAG2 pins become floating outputs.
When the FLAG1 and the FLAG2 pins are
optioned as 3HzB and 3Hz output, they will
be alternately output at a 3Hz rate.
♦
Level trigger
When one of the eight keys i s pressed and
held down, the corresponding group will
keep playing. Once the pressed key is released, the group will not stop till the included sections are all completed.
•
6Hz/6HzB flash
When voices are playing, the FLAG1 pin
outputs a 6Hz signal to drive an LED. The
signal is active low , 25% duty . Once the voice
output is terminated, the FLAG1 pin becomes a floating output. When the FLAG1
and the FLAG2 pins are optioned as 6HzB
and 6Hz outputs, they will be alternately
output at a 6Hz rate.
7 5th May ’98
•
Busy output
When a voice group is playi ng, the ou tputs of
both FLAG1 and FLAG2 are turned low , indicating that the chip is busy.
In addition to the above-stated output signals,
FLAG1 can also generate one of the following
signals by code option:
•
Voice indicator output
FLAG1 is active low when voices are playing.
FLAG1 is also turned low when a voice section
is output. The brightness of FLAG1 varies
with the volume. FLA G1 becomes floating after the silence section is output or the voice
output is terminated.
HT813D0
Volume control
The function of the volume control can be set by
mask option. A code is written in th e function
table for the purp ose of contro lling the vol ume
of each section output after the volum e control
function is chosen. There are two volume options, namely, full range and half range.
AUD
The AUD pi n is a PM OS op en dra in stru cture.
It outputs voice signals to drive a speaker
through an external NPN transistor when the
chip is active. However, the AUD pin becomes a
floating output when the chip is in the standby
state.
The 8050 type trans istor with h
ommended for an output driver.
≅150 is rec-
FE
•
End-pulse output
When the voice output is completed, the
FLAG1 pin outp uts an active low pulse. The
pulse width can be programmed depending on
the customer’s requirements.
The FLAG1 as well as FLAG2 pins are both
floating outputs when the chip is in the
standby state.
8 5th May ’98
Application Circuits
General application
HT813D0
9 5th May ’98
Parallel application
Cascade or external driving
HT813D0
Power- on play (one shot)
10 5th May ’98
Power-on play (continuous)
With volume adjustment
HT813D0
11 5th May ’98
HT813D0
Push-pull output
To prevent the speaker and driver transistor from damage due to excess powe r dissipation which
results from a high voltage power supply (4.5V~5.5V), the following push-pull output stage is
recommended.
Coupling to power amplifier
Standard Item List
Item Name R
HT813D1 Brick Game 460kΩ 600µs50kΩNote 1
Note 1: KEY1: Sequential
KEY2~KEY8: Level-trigger
Key Debounce
OSC
Time
12 5th May ’98
Pull-High
Resistor
Key Function
Loading...