ROHM Speech Synthesis ICs Aim to Improve Driver Safety

In the automotive industry, advanced technologies are not only enhancing the driving experience but also playing a critical role in safety. One such technology is speech synthesis, which has gained significant importance, especially in the context of electric and hybrid vehicles. 

 

Illustration of automotive speech synthesis. Image used courtesy of Rohm

 

ROHM Semiconductor has released a family of speech synthesis ICs for automotive applications. 

 

Speech Synthesis for Automotive

In today's automotive industry, the integration of speech synthesis technologies has become increasingly essential, driven by the need for enhanced safety, user experience, and compliance with evolving regulatory standards. 

 

Speech synthesis for automotive use. Image used courtesy of ROHM Semiconductor

 

Pedestrian safety presents a critical need for speech synthesis in vehicles, especially in electric and hybrid vehicles. These vehicles, known for their quiet operation, pose a unique risk to pedestrians, cyclists, and other road users who traditionally rely on auditory cues to gauge vehicle presence and movement. Speech synthesis systems, particularly in Acoustic Vehicle Alerting Systems (AVAS), provide a vital solution by generating human-like sounds that alert pedestrians to an approaching or idling vehicle. This is particularly useful in urban and residential areas where pedestrian traffic is high.

Speech synthesis also plays a pivotal role in enhancing the in-vehicle experience. Modern vehicles are equipped with numerous systems, such as navigation, infotainment, and driver assistance technologies. Speech synthesis enables these systems to communicate with the driver and passengers, providing instructions, information, and alerts in a clear, understandable manner. 

 

The ML22120TB Speech Synthesis IC

ROHM Semiconductor’s ML22120xx family consists of a series of speech synthesis ICs for automotive applications. One notable product in this group is the ML22120TB, a speech synthesis LSI for automotive applications. 

Notably, this device integrates a serial flash memory interface compatible with sound data playback functions, commonly called sound generators. The core technology of these LSIs is grounded in their ability to dynamically control pitch and volume through specialized setting registers, offering customizable sound outputs for various applications.

 

Application circuit of the ML22120TB. Image used courtesy of ROHM

 

At the heart of the ML22120 series is a 16-bit straight pulse code modulation speech synthesis algorithm, which can store up to 64 phrases with a maximum flash memory capacity of 128 Mbits. The series offers a flexible sampling frequency range from 8.0 to 48.0 kHz.  Depending on the selected sampling frequency, the maximum sound production time can range from 174 seconds at 48.0 kHz to an extended 1,048 seconds at 8.0 kHz​.

The series also incorporates a 16-bit digital-to-analog converter and a low-pass filter, enhancing the quality of the synthesized sound. It supports a line amplifier output capable of driving 10 kΩ, making it suitable for direct connection to an external speaker amplifier. For control and communication, it includes clock synchronous serial and I2C interfaces (slave), with a master clock frequency of either 4.096 MHz or 4.000 MHz. 

These features make the ML22120xx series particularly well-suited for automotive applications where customized, clear, and reliable sound output is critical.

 

Improving Automotive Safety

The launch of ROHM’s speech synthesis ICs is noteworthy because it enhances the safety features of vehicles and contributes to an improved driving experience. These ICs could improve the future of automotive safety in electric and hybrid systems.