Speech Synthesis LSI Special Site

Automotive-grade Speech Synthesis LSI "ML22530" Video Supporting communication between people and vehicles with sound, realizing a safe, secure, and comfortable driving environment.

Introduction of Speech Synthesis LSI "ML22530 Family"

08:29

Introduction of Speech Synthesis LSI "ML22530 Family"

08:29

LAPIS Technology's Speech Synthesis LSI
5 points! "Easy" "Fast" "High sound quality" "Multifunctional" " Space saving"
A high-quality audio solution can be immediately realized on your device.

Lapis' Speech Synthesis LSI is easy: Audio can be output simply by sending simple commands from the microcontrollers.
LAPISs speech synthesis LSI is fast: Playback starts in about 5 ms after inputting a playback request. Easy control by simply inputting playback commands or setting registers.
Lapis Speech Synthesis LSI has high sound quality: Sampling frequency up to 48kHz (middleware is typically 8-12kHz).

Built-in high-performance DAC and filter.
Lapis' Speech Synthesis LSI saves space: Built-in from voice synthesis engine to speaker amplifier, almost no external parts are needed.

Reduce the risk of a failure rate rise by reducing the number of parts.

Lapis' Speech Synthesis LSI has multiple functions: Built-in various functions related to sound. Equipped with fail-safe functions such as abnormal sound detection for vehicles.

Multiple voice simultaneous utterance function.

Editing ROM function etc.

Please leave "sound making" Lapis Technology's speech synthesis solution provides one-stop support
for " voice recording"," melody creation"," sound editing"," data writing, and shipping".
You can also select a sound from a "rich audio library"without having to make a sound.
We easily realize "I want to make a sound!".

Customer work
STEP 1 Select LSI by application and ROM capacity: Select an LSI according to your requirements such as application, playback time, required functions, etc.
STEP 2 Create a phrase list

Our and / or customer work
STEP 3 Original sound creation: Select from studio recording, melody / sound effect creation, and audio library
STEP 4 Voice analysis and editing: Carving, volume control, splicing, equalizing
STEP 5 Complete sound data, write to ROM, and ship

High reliability -40°C to +105°C Suitable for automotive applications "AEC-Q100 Grade 2 and products meeting the requirements of the standard""Manufacturing in IATF 16949 certified facility"
Supporting automotive applications for customers in harsh environments.

AEC-Q100 qualified product suitable
for automotive applications

LAPIS Technology offers AEC-Q100 qualified products and products that meet the requirements for automotive applications.
AEC-Q100 is an industry standard for automotive reliability of automotive integrated circuits (ICs) and is widely adopted as a standard for automotive electronic components in Europe and the United States.
The AEC-Q100 has different grades according to the operating temperature range of the IC. LAPIS Technology's Speech Synthesis LSI for automotive applications belongs to [Grade 2] of -40°C to +105°C.

AEC-Q100 test Items

Environmental stress test
Accelerated life test
Package assembly maintenance test
Die level reliability test
Confirmation of electrical characteristics
Cavity-package test, etc

Manufacturing in a new IATF 16949 certified facility for the automotive industry

Speech Synthesis LSIs manufactured by LAPIS Technology are manufactured at IATF 16949 certified facilities.
The International Automotive Task Force (IATF) is a new quality management system standard for the automotive industry that replaces ISO / TS 16949.
Manufactured in IATF 16949 certified facility, we will provide stable and better quality products.

All production sites acquired IATF 16949 certification

LAPIS Technology Quality

Quality management system: ISO9001, standard for the automotive industry: IATF16949
Ensuring high reliability and quality stability.

Features Automotive-grade products are equipped with "various fail-safe functions"
Monitors various LSI abnormalities and further enhances the safety of your automotive applications.

Feature 1Reduction of software development man-hours and costs

Hardware integration of the sound functions ensures quick response and sound quality of the vehicle approach alert sound even under heavy system load through simple command control.
In addition, verification related to the playback of the sound of approaching vehicle alerts is no longer required, contributing to a reduction in man-hours and costs for software verification.

Applicable models : All models of in-vehicle speech synthesis LSIs

Feature 2Easy adjustment of vehicle approach warning sound

Pitch control (4 independent channels), volume control (4 independent channels), and frequency adjustment using a 5-band equalizer make it easy to adjust the vehicle approach warning sound without changing the sound source.

Applicable models : ML22120

Feature 3Equipped with a variety of fail-safe functions

Multiple built-in fail-safe functions, such as detecting playback abnormalities, speaker pin short-circuits, internal LSI temperature, supply voltage drops, and MCU command transfer errors, make it possible to monitor for a variety of LSI failure modes.

Applicable models : All models of in-vehicle speech synthesis LSIs
However, the fail-safe function installed differs depending on the model.

Output becomes "H" when ground faults occur between the SP pins (output normally "L")

Output becomes "H" at high temperature (output normally "L")

If the commands are different two times, the output becomes H (output normally "L")

Feature 4Sound data can be rewritten even after customer's product shipment

[Flash memory] and [External flash memory] can rewrite sound data via the SPI I/F of the microcontroller.
sound data can be rewritten in the field after the customer's product is shipped.

Applicable models : ML22Q532,ML22Q533,ML22530
ML22120

Feature 5+105°C guaranteed even with built-in speaker amplifier

Automotive-grade Speech Synthesis LSIs are offered to feature a guaranteed operating temperature range of –40°C to +105°C. Integrated memory and speaker amp, together with support for high-temperature +105℃ operation, ensure worry-free use in automotive applications.
The exposed backside die pad enables heat dissipation through the substrate, making it possible to suppress temperature rise within the LSI (including the speaker amp) and ensure stable performance even at an ambient temperature of +105°C.

Applicable models : ML22Q532,ML22Q533,ML22530
ML22Q573,ML22Q553,ML22572,ML22573,ML22594
ML22Q274,ML22Q284,ML22Q294

LSI lineup

Comparison of memory capacity and type and playback time of each series

ROM(bit)
0.692M
2M
4M
6M
8M
16M
32M
128M

For automotive applications

flash : ML2253x series [Writable by CPU command I/F (in SPI)]
flash : ML2257x series
flash : ML222xx series
MaskROM : ML2253x series
External memory : ML2253x series [Writable by CPU command I/F (at SPI)]
External memory : ML225xx series
External memory : ML22120 [Writable by CPU command I/F (at SPI)]

For industry and applications

Long playback

flash : ML226xx series [Writable by CPU command I/F (at SPI)]
P2ROM : ML228xx series
P2ROM : ML227xx series

Short playback

flash : ML2256x series
flash: ML222xx series
MaskROM : ML2256x series

Supports external memory

External memory : ML226xx series [Writable by CPU command I/F (at SPI)]
External memory : ML224xx series

27sec.
Playtime
1,044sec.
(17min. 24sec.)

Ideal for
sound effects
Ideal for
guidance

Automotive-grade Speech Synthesis LSI (Flash memory, Mask, External)

	Operating voltage (V)	Operating frequency (MHz)	Oscillator	Operating temperature (°C)	ROM capacity (bit)	Number of phrases	Maximum playback time	CPU I/F	SP amplifier output(W)/ class	Mixing number (internal) (ch)	Other	PKG	For automotive^*7	Sample and Buy
ML22120	2.7 to 3.6	4.096 4.000	Built-in External	−40 to ＋105	External maximum 128M	64	17min	I²C / Clock syncronous serial	-	4	Pitch and volume control 5band Equalizer Failure detection	TQFP32 WQFN32 WQFN24	〇
ML22Q532	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋105	Flash 2M	4096	99sec *1	I²C / Clock syncronous serial	1.0/ AB-class	4	Rewrite Flash from MCU *5 Serial Audio Interface Failure detection	TQFP48	〇
ML22Q533	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋105	Flash 4M	4096	202sec *1	I²C / Clock syncronous serial	1.0/ AB-class	4	Rewrite Flash from MCU *5 Serial Audio Interface Failure detection	TQFP48	〇	Buy
ML22530	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋105	External maximum 128M	4096	109min *1	I²C / Clock syncronous serial	1.0/ AB-class	4	Rewrite Flash from MCU *5 Serial Audio Interface Failure detection	TQFP48	〇	Buy
ML22Q274	2.0 to 5.5	4.096	Built-in	−40 to ＋105	Flash 692K *6	30/62	28sec *2	Clock syncronous serial	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20	〇
ML22Q284	2.0 to 5.5	4.096	Built-in	−40 to ＋105	Flash 692K *6	30	28sec *2	Stand-alone	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20	〇
ML22Q294	2.0 to 5.5	4.096	Built-in	−40 to ＋105	Flash 692K *6	30/62	28sec *2	I²C	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20	〇
ML22572	2.7 to 5.5	4.096	External	−40 to ＋105	Mask 2M	1024	99sec *1	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30	〇
ML22573	2.7 to 5.5	4.096	External	−40 to ＋105	Mask 4M	1024	202sec *1	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30	〇
ML22Q573	2.7 to 5.5	4.096	External	−40 to ＋105	Flash 4M	1024	202sec *1	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30	〇
ML22Q553	2.7 to 5.5	4.096	External	−40 to ＋105	Flash 4M	1024	202sec *1	Clock syncronous serial	1.0/ AB-class	4	Speaker terminal Short circuit detection function	SSOP30	〇
ML22594	4.5 to 5.5	4.096	External	−40 to ＋105	Mask 6M *4 External maximum 128M	1,024 *5 ( Built-in 512/ External 512 )	Built-in 304sec 1 External 109min 3	Clock syncronous serial	1.0/ AB-class	4	Speaker terminal and short circuit detection function	SSOP30	〇

*1 : Maximum playback time when sampling frequency is 6.4kHz with HQ-ADPCM.
*2 : Maximum playback time when sampling frequency is 6.4kHz in ADPCM2.
*3 : When using an external memory (Max. 128Mbit), the maximum playback time when the sampling frequency is 6.4kHz with HQ-ADPCM.
*4 : Built-in Mask ROM is 6Mbit and external memory (Max.128Mbit) can be connected.
*5 : MTotal value of 512 phrases built-in Mask and 512 phrases in external memory.
*6 : For clock synchronous serial.
*7 : 〇Is a vehicle compatible product. 〇 Can provide AEC-Q100 qualified products.

Long Playback (Flash memory, P2ROM™: read-only memory for short delivery time)

	Operating voltage (V)	Operating frequency (MHz)	Oscillator	Operating temperature (°C)	ROM capacity (bit)	Number of phrases	Maximum playback time	CPU I/F	SP amplifier output (W) / class	Mixing number (internal) (ch)	Other	PKG	Sample and Buy
ML22Q623	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 4M	4096	202sec *3	Clock syncronous serial	1.0/AB-class、 D-class	4	Rewrite Flash from MCU Failure detection	TQFP32 WQFN32
ML22Q624	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 8M	4096	406sec *3	Clock syncronous serial	1.0/AB-class、 D-class	4	Rewrite Flash from MCU Failure detection	TQFP32 WQFN32
ML22Q625	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 16M	4096	14min *3	Clock syncronous serial	1.0/AB-class、 D-class	4	Rewrite Flash from MCU Failure detection	TQFP32 WQFN32	Buy
ML22Q626	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 32M	4096	27min *3	Clock syncronous serial	1.0/AB-class、 D-class	4	Rewrite Flash from MCU Failure detection	TQFP32 WQFN32
ML22Q663	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 4M	4096	202sec *3	I²C	1.0/AB-class、 D-class	4	Failure detection	TQFP32 WQFN32
ML22Q664	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 8M	4096	406sec *3	I²C	1.0/AB-class、 D-class	4	Failure detection	TQFP32 WQFN32
ML22Q665	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 16M	4096	14min *3	I²C	1.0/AB-class、 D-class	4	Failure detection	TQFP32 WQFN32	Buy
ML22Q666	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋70	Flash 32M	4096	27min *3	I²C	1.0/AB-class、 D-class	4	Failure detection	TQFP32 WQFN32
ML22823	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 4M	4,096 *2	161sec *1	Clock syncronous serial	0.7/ AB-class	2	―	SSOP30
ML22824	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 8M	4,096 *2	325sec *1	Clock syncronous serial	0.7/ AB-class	2	―	SSOP30
ML22825	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 16M	4,096 *2	11min *1	Clock syncronous serial	0.7/ AB-class	2	―	SSOP30
ML22863	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 4M	4,096 *2	161sec *1	I²C	0.7/ AB-class	2	―	SSOP30
ML22864	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 8M	4,096 *2	325sec *1	I²C	0.7/ AB-class	2	―	SSOP30
ML22865	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 16M	4,096 *2	11min *1	I²C	0.7/ AB-class	2	―	SSOP30
ML22724	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 8M	4,096 *2	325sec *1	Clock syncronous serial	0.7/ AB-class	1	Speech-speed and pitch conversion	SSOP30
ML22725	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 16M	4,096 *2	11min *1	Clock syncronous serial	0.7/ AB-class	1	Speech-speed and pitch conversion	SSOP30
ML22764	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 8M	4,096 *2	325sec *1	I²C	0.7/ AB-class	1	Speech-speed and pitch conversion	SSOP30
ML22765	2.7 to 3.6 or 4.5 to 5.5	4.096	External	−40 to ＋85	P2ROM^TM 16M	4,096 *2	11min *1	I²C	0.7/ AB-class	1	Speech-speed and pitch conversion	SSOP30

*1 : Maximum playback time when sampling frequency is 6.4kHz in ADPCM2.
*2 : 1024 phrases (1 bank) × 4 banks
*3 : Maximum playback time when sampling frequency is 6.4kHz with HQ-ADPCM.
*4 : 256 phrases (1 bank) × 2 banks.
*5 : 256 phrases (1 bank) × 4 banks.

Short Playback (Flash memory, MaskROM)

	Operating voltage (V)	Operating frequency (MHz)	Oscillator	Operating temperature (°C)	ROM capacity (bit)	Number of phrases	Maximum playback time	CPU I/F	SP amplifier output (W)/class	Mixing number (internal) (ch)	Other	PKG	For automotive^*3
ML22Q234	2.0 to 5.5	8.192	Built-in	−40 to ＋85	Flash 676K *2	30/62	34sec *1	Clock syncronous serial	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20
ML22Q244	2.0 to 5.5	8.192	Built-in	−40 to ＋85	Flash 676K *2	30/62	34sec *1	Stand-alone	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20
ML22Q254	2.0 to 5.5	8.192	Built-in	−40 to ＋85	Flash 676K *2	30/62	34sec *1	I²C	1.0/ D-class	1	Disconnection and short circuit detection function	TSSOP20
ML22Q374	2.0 to 5.5	4.096	Built-in	−40 to ＋85	Flash 692K	30	28sec *2	Clock syncronous serial	1.0/ D-class	1	Disconnection Short circuit detection	TSSOP20 SSOP16	〇
ML22Q394	2.0 to 5.5	4.096	Built-in	−40 to ＋85	Flash 692K	30	28sec *2	I²C	1.0/ D-class	1	Disconnection Short circuit detection	TSSOP20 SSOP16	〇
ML22562	2.7 to 5.5	4.096	External	−40 to ＋85	Mask 2M	1024	99sec	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30
ML22563	2.7 to 5.5	4.096	External	−40 to ＋85	Mask 4M	1024	202sec *1	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30
ML22Q563	2.7 to 5.5	4.096	External	−40 to ＋85	Flash 4M	1024	202sec *1	Clock syncronous serial	1.0/ AB-class	4	Fail safe	SSOP30

*1 : Maximum playback time when sampling frequency is 6.4kHz with HQ-ADPCM.
*2 : Maximum playback time when sampling frequency is 6.4kHz in ADPCM2.
*3 : 〇Is a vehicle compatible product. 〇 Can provide AEC-Q100 qualified products.

Speech Synthesis LSI for External Memory

	Operating voltage (V)	Operating frequency (MHz)	Oscillator	Operating temperature (°C)	ROM capacity (bit)	Number of phrases	Maximum playback time	CPU I/F	SP amplifier output (W) / class	Mixing number (internal) (ch)	Other	PKG	Sample and Buy
ML22620	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋85	External maximum 128M	4096	109min *2	Clock syncronous serial	1.0/AB-class、 D-class	4	Rewrite Flash from MCU Failure detection	TQFP32 WQFN32	Buy
ML22660	2.7 to 3.6 or 3.3 to 5.5	4.096 4.000	Built-in External	−40 to ＋85	External maximum 128M	4096	109min *2	I²C	1.0/AB-class、 D-class	4	Failure detection	TQFP32 WQFN32	Buy
ML22420	2.7 to 5.5	4096	External	−40 to ＋85	External maximum 128M	1024	87min *1	Clock syncronous serial	0.7/AB-class	4	―	SSOP30
ML22460	2.7 to 5.5	4096	External	−40 to ＋85	External maximum 128M	1024	87min *1	I²C	0.7/AB-class	4	―	SSOP30

*1 : Maximum playback time when sampling frequency is 6.4kHz in ADPCM2.
*2 : When using an external memory (Max. 128Mbit), the maximum playback time when the sampling frequency is 6.4kHz with HQ-ADPCM.

Application : Automotive

EV / HV AVAS

EV / HV startup sound

EV / HV pseudo engine sound

Voice of parking
assist system

Warning sound
(seat belt / half door)

Operation sound
(Blinkers)

Industrial equipment, consumer products

Elevator: voice guidance

Card reader:
voice guidance

Security: voice guidance

Home appliances:
voice guidance

Various alarms:
warning sound

Housing equipment:
Operation sound

Remote control:
operation sound

Home appliances:
operation sounds

Development support tool

Sound device control kit "SDCK"

This is a development tool that can perform everything from the creation of voice data to the writing of voice data into the voice synthesis LSI, to the trial listening.

* Two types, "SDCK" and "SDCK3", are available depending on the target LSI. For details, refer to “Lineup”.

Software

Speech LSI Utility

A tool to create voice data for speech synthesis LSI.
The created audio data can be simulated on a PC.

Wave Editor

A tool for editing the waveform of original sound data.

SDCB Controller

This is a tool for writing data to the Speech Synthesis LSI on the reference board via the following hardware SDCB and for listening.
Commands to be sent to the Speech Synthesis LSI can be easily operated in GUI format.

Hardware

Sound Device Control Board "SDCB2" "SDCB3"

This tool is to write data to the Speech Synthesis LSI on the reference board and to audition according to the operation of the SDCB Controller.

The reference board is replaceable and supports the Speech Synthesis LSI of the connected reference board.

* Writing for mass production is not covered by the warranty. For mass production, please use the writing shipment at the LAPIS Technology factory or use an on-board writer.

* Two types, "SDCB2" and "SDCB3", are available depending on the target LSI. For details, refer to “Lineup”.

Reference board (sold separately)

It is prepared according to each Speech Synthesis LSI.
Each board is equipped with a speaker jack, so you can easily listen to it just by connecting a speaker.
It can also be used when connected to your system board.

Possible to purchase online

LSI

Reference board

Sound Device Control Kit

Contact us Please contact us here for purchasing LSIs and development tools, technical questions, etc.

Inquiry

Contact form

Speech Synthesis LSI link

Automotive-grade

Automotive-grade Speech Synthesis LSI with 4-channel mixing, built-in speaker amplifier, and guaranteed operating temperature +105°C. AEC-Q100-qualified products can be provided. As the fail-safe function, it is equipped with functions necessary for in-vehicle use such as detection of internal temperature of LSI, supply voltage drop detection, and command transfer error detection.

Long Playback

Long playback Speech Synthesis LSI ML22Q6xx has a large-capacity 4 to 32 Mbit Flash memory with rewritable voice data. The ML228xx and ML227xx incorporate LAPIS Technology's proprietary memory P2ROM capable of short TAT from 4Mbit to 16Mbit. Both are ideal for long playback such as voice operation guidance.

Short Playback

Short playback Speech Synthesis LSI has a built-in ROM of 692Kbit to 4Mbit, and speech playback is possible simply by inputting a simple command from the MCU.
This LSI prepares various input interfaces to meet the needs of a wide range of customers.
The speech playback function can be easily realized by simply adding this LSI to the customer's system.

Supports External Memory

External memory compatible speech synthesis LSIs ML226xx and ML224xx can connect up to 128Mbit serial memory externally. Ideal for long sound playback.

Speech playback MCUs (8bit)

High-quality sound reproduction function and high-output speaker amplifier are integrated into one chip. White goods can easily have a sound playback function. We provide speech playback functions from dry cell powered devices such as thermometers and healthcare devices to security devices such as fire alarms, and AC devices such as homes and home appliances.

Development support system

A series of evaluations are possible from hardware (SDCB, Reference board) and software (Speech LSI Utility). Starter kits are available for the development of audio playback 8-bit microcontrollers.

Contents list

Company names, product names, and service names appearing on this website are trademarks or registered trademarks of their respective companies.

ROHM-Header

speech-lsi

A lineup of highly reliable in-vehicle devices with various fail-safe functions. Speech Synthesis LSI series

A lineup of highly reliable in-vehicle devices with various fail-safe functions. Speech Synthesis LSIseries