L&H.tts2000/T
by Lernout & Hauspie Speech Products

Software Overview

Lernout & Hauspie's Text-To-Speech solution converts any computer-readable text into intelligible 
human-sounding synthetic speech. The technology is based on parameterized segment concatenation 
algorithms, where basic human voice segments such as diphones, triphones, and tetraphones are 
stored and used to convert the text into speech. In-depth language specific linguistic knowledge is 
used for syntactic and lexical analysis.

Features and Benefits

•	Unlimited vocabulary
•	Reduced memory requirements
•	Customizable vocabulary tool
•	Mutiple channels
•	Transplanted prosody tool

Processor and System Specifications

•	Devices supported: TMS320C31
•	Algorithm category: Text-to-Speech
•	Requirements:
Program memory (in words): 110 kbytes
Data memory (in words): 370–560 kbytes
Processor loading: 4 MIPS
•	Sampling rate: 8 kHz
•	Languages supported: English (US and British), French, German, Spanish, Dutch, Italian

Availability

•	Now

Company Background and Contact Information

Lernout & Hauspie Speech Products is the international leader in the development of speech 
technology for commercial products in four markets: Telecommunications, Computers & Multimedia, 
Automotive Electronics, and Business Electronics. The company offers more speech processing 
products, for more worldwide markets, than any other company. Its integrated products are built on 
its core speech recognition, text-to-speech, and speech-coding technologies. Available in multiple 
languages and multiple platforms.

Contact:	Scott Garlick
		Sint-Krispijnstraat 7
		8900 Ieper, Belgium
		32-2-456-0500
		Fax: 32-2-460-0172
Contact:		Burr Tupper
		800 West Cummings Park 
		Suite 3100
		Woburn, MA 01801
		USA
		(617) 932-4118
		Fax: (617) 932-9209

Contact:		G. K. Chen
		RM 9, 4F, No. 6, Lane 609
		Section 5
		Chung Hsing Rd, SanChung
		Taipei Hsien, Taiwan R.O.C.
		886-2-999-2415
		Fax: 886-2-999-2416

		WWW homepage address: 
		www@lhs.com