S1V3S344 Voice Guidance LSI Overview The S1V3S344 is an LSI incorporating built-in flash memory for voice data and featuring pin compatibility with existing S1V3034x Series devices*. It features high-compression, high-quality audio decoding functions, built-in voice data flash memory, and a DA converter, making it ideal for use in voice guidance products. The voice data creation tool for EPSON voice guidance LSI allows easy creation of high-quality voice data without the need for studio recording. All functions are controlled by commands via a serial interface for easy addition to any existing system incorporating a host. The S1V3S344 will help evaluate voice output and help users reduce time-to-market for products incorporating built-in voice guidance functions. * External parts differ from S1V3034x Series devices. Features Audio playback Host interface - High-compression, high-quality audio decoder - Clock synchronized serial interface, (proprietary Epson data format) supporting UART and I2C - Bitrate: 40 kbps, 32 kbps, 24 kbps, 16 kbps - Command control - Sampling rate: 16 kHz High-quality 16-bit DA converter Sequencer function (phrase interval setting) - Sampling rate (f ): 16 kHz s - Sequence setting for up to 64 phrases - Input bits: 16 bits (unlimited combinations) Clock - Variable phrase interval delay setting: 0 ms or 20 ms to - Clock input: 32.768 kHz or 12.288 MHz 2,047 ms (in 1 ms steps) - Crystal oscillator: 32.768 kHz 1 * Voice data built-in flash memory Package - Incorporates the following memory for voice data - QFP-52pin (10 mm 10 mm) 0.65 mm pin pitch 512 kbytes (approx. 4 minutes/16 kbps) Power supply voltage Erase/write cycles: 10,000 cycles (typ.) - 5.0 V 0.5 V (single power supply) - 3.3 V 0.3 V (single power supply) 1,000 cycles (min) Data retention: 10 years (min) Standard application system The S1V3S344 standard application system is configured as shown in the diagram below. The S1V3S344 is command-controlled by the host using a messaging protocol via the serial interface. Controlled by commands sent from the host via the serial interface after power-on resetting, the S1V3S344 outputs voice audio while internally decoding and processing internal or streamed (via host command transfer) compressed audio data. Host Serial interface Command controller (SPI/UART/I2C) Voice decoder DA converter Speaker Clock 32.768 kHz or 12.288 MHz Crystal oscillator 32.768 kHz Built-in voice flash memory 512 KB Power supply voltage (approx. 4 mins) 4.5 V to 5.5 V 3.0 V to 3.6 V (single power supply) SRAM ROM SEIKO EPSON CORPORATION S1V3S344 Development Tools - Evaluation board 2 * - Voice data creation tool - Sample programs Voice data creation tool overview Step 1: Voice guidance registration Step 2: Pronunciation editing Step 3: Phrase editing Step 4: ROM data production Supported languages : English, Japanese, Korean (all female voices) In planning stage : Spanish, Chinese *1 The flash memory technology used in this product is used under license of Silicon Storage Technology, Inc. in the USA. *2 The audio data creation tool uses PENTAXVoiceTex voice synthesis technology.VoiceTex is a registered trademark of Voiceware Co., Ltd. Companies and product names included here are registered trademarks or trademarks. SEIKO EPSON CORPORATION 2009, All rights reserved. Document Code: 411676800 Phone: +81-42-587-5313 FAX: +81-42-587+5116 First Issue March 2009 in JAPAN D