Embedded Text-to-Speech
BUSINESS PROBLEM: You require speech synthesis technology with a relatively small footprint capable of running in a wide variety of products across multiple operating systems and hardware platforms.
TECHNOLOGY SOLUTION: An Embedded Text-to-Speech Engine
This technology allows for speech applications that are housed on the devices where they will be used. Voice technology can be deployed rapidly because there is nothing for the user to download or install.
Product: AT&T Natural Voices Text-to-Speech
Delivery Method: Embedded software available through Wizzard Software.
Pricing: Varies with vertical markets, volume, and licensing models. For an IVR, Wizzard can license for less than $100 per port for high-volume customers. Wizzard will also license audio generated by the software to Web sites starting at $5,500 .
Functionality: AT&T Natural Voices supports Linux, Windows, and Macintosh platforms. It has two different engines: one that handles one session at a time, and a server version that can handle several at a time. It offers 20 voices in eight languages at 8 kHz and 16 kHz sampled rates.
Business Benefits: Eliminates the cost of prerecording and allows for a dynamic spoken interface, independent of preprescribed scenarios.
Contact: Wizzard Software at 1-954-678-4155; www.wizzardsoftware.com.
Product: Loquendo TTS
Delivery Method: Embedded software.
Pricing: For consumer market navigation devices, licenses start at $4 per unit. Prices vary significantly for different vertical markets and with volume.
Functionality: Loquendo TTS is available in 26 languages with 62 voices. Platforms supported are Windows Mobile 6.0 and 5.0, CE 6.0 and 5.0, Windows XP Embedded and TabletPC, VXWorks, Linux, Symbian OS S60, QNX, and iPhone; the Loquendo API (C/C++ and Compact .NET Framework), SAPI 5, and W3C SSML 1.0 interface are also supported. Loquendo TTS can also read mixed-language texts. A single TTS voice can read texts in several major languages while maintaining its native accent, or the voice can be changed with each language change. Loquendo provides tools, such as Voice Creator and Lexicon Manager, to allow for greater customization.
Business Benefits: Once installed, Loquendo TTS allows scripts to be rapidly deployed and eliminates the need for prerecording. Prerecorded messages, however, can be seamlessly integrated.
Contact: Loquendo at +39-011-291-3111; www.loquendo.com.
Product: SVOX Automotive Pro
Delivery Method: Embedded software.
Pricing: Automotive Pro licenses cost about $12 per unit (with a volume of 10,000 licenses per year). Special prices apply for bundles with SVOX speech recognition products. SVOX can sell only to high-volume customers.
Functionality: SVOX Automotive Pro supports 21 languages, including major European, North American, and Asian languages. It consists of the SVOX engine and a number of lingware packages. It requires approximately 7 megabytes of RAM and 12 megabytes of ROM (Flash, hard disk, or DVD) per included language. SVOX can also provide less memory-intensive configurations and extended language libraries of up to 300 megabytes.
Business Benefits: Automotive Pro enhances the on-board features of any vehicle. It provides a dynamic interface capable of becoming the voice behind driving directions, user interfaces, etc.
Contact: SVOX at +41-43-544-0600; www.svox.com.