Fonix Speech has updated its automatic speech recognition (ASR) software that runs on Linux, among other OSes. The company claims that VoiceIn SE 4.1 boosts recognition by 24 percent, to accuracy rates as high as 98 percent, and offers significantly improved recognition accuracy in “noise-saturated” environments.

VoiceIn SE 4.1 targets PDAs and mobile phones, electronic dictionaries, medical devices, industrial data collection and inventory applications, multimedia players, and automotive telematics systems.

A key addition to version 4.1, according to Fonix, is a speech-analysis module that enables the development of applications that provide feedback to end users, to help them improve their pronunciation of foreign words and phrases. The module “utilizes significantly more phonetic, linguistic, and prosodic information about the speech signal, [helping] end users learning to speak foreign languages to practice proper pronunciation,” Fonix said. The technology is available for all VoiceIn-supported languages (listed below).

Fonix VoiceIn 4.1 supports Linux, Win32, Windows CE, Windows Mobile, QNX, and Symbian. Supported processor architectures include ARM, XScale, Freescale i.MXL, TI OMAP, Renesas SH3/SH4, MIPS, PowerPC, and others. Additionally, the software is available in numerous languages, including U.S. and U.K. English, Canadian and European French, Castilian and Latin American Spanish, German, Japanese, Swedish, Italian, and Korean.

Availability

VoiceIn 4.1 is shipping now, direct from the company. Pricing was not disclosed.

 
This article was originally published on LinuxDevices.com and has been donated to the open source community by QuinStreet Inc. Please visit LinuxToday.com for up-to-date news and articles about Linux and open source.