Fonix Speech FAQ

What does Fonix Speech, Inc. do?
Fonix provides automatic speech recognition (ASR), text-to-speech (TTS), and embedded speech interfaces for mobile devices, handheld electronic products, video game consoles, data base systems and processors.
What are Fonix’s current service and product offerings?
Fonix provides original equipment manufacturers (OEMs) and original device manufacturers (ODMs) with cost-effective speech solutions to enhance devices and systems for added convenience and functionality.
What industries does Fonix serve?
Fonix currently offers voice technology for mobile/wireless devices; electronic games and game consoles, toys and appliances; computer telephony systems and server applications; the assistive and language learning markets; and vehicle telematics.
What consumer products feature Fonix ASR and/or TTS technologies?

Games

Fonix VoiceIn® Game Edition is our voice command software available for cross-platform game developers who want to employ speech interfaces in games.

Now creators of cross-platform games can use the same voice command software on multiple platforms:

  • Nintendo Wii™
  • Xbox®
  • Xbox 360®
  • PlayStation® 2
  • PlayStation® 3
  • PC

Fonix VoiceIn software is optimized for game development where memory and processing power are at a premium. Product editions include VoiceIn Standard Edition, VoiceIn Game Edition, VoiceIn Karaoke Edition, and VoiceIn Phonetic Edition.

Fonix VoiceIn voice command software is also available to game developers in multiple languages:

  • US and UK English
  • German
  • French
  • Spanish
  • Italian
  • Korean
  • Japanese

Assistive market

Fonix speech solutions make everyday life a lot easier for people who are blind or have visual, vocal or mobility impairments. FonixTalk is the assistive industry's premier TTS engine, offering nine highly intelligible voices and multiple languages. Many users rely on FonixTalk to read their email, the daily news or other documents, or to function as their voice to the outside world. In addition, we have expanded our solutions for the assistive market to incorporate the full line of Fonix TTS offerings, including high-quality concatenated TTS and high-recognition-rate ASR. These offerings meet assistive users' needs whether someone has a learning disability requiring a more natural voice or a disabled user who requires voice-activated input methods.

Language learning

Fonix speech solutions are particularly useful for non-English speakers who are learning the language. In the speech-enabled language learning market, we have capitalized on FonixTalk's small memory footprint and high intelligibility and Fonix VoiceIn's high-recognition-rate capabilities. Speech-enabled language learning is an emerging market, especially in Asia. Manufacturers are selling handheld electronic dictionaries that allow individuals to speak a word in their native language (like Korean) and have the text read back to them in English. Educational electronic dictionary devices are growing in popularity in China and are expected to exceed a market volume of over 500,000 units per quarter. These channels expand our ability to serve millions of individuals while generating revenue on already existing technologies that can be diverse in their final application. Our goal is to become the primary supplier of speech solutions for OEMs providing language learning devices and systems.

PDA, PC and electronic devices

Fonix speech solutions apply the intuitive use of voice interfaces to tasks that users perform everyday. Many of these solutions are appropriate for multiple markets -- assistive, mobile and wireless, and business and home users. Fonix technologies enable users to interact naturally with electronic devices. With Fonix speech solutions, users may listen to documents of any length, have email read aloud, access programs and launch applications by speaking.

What speech applications use Fonix technologies?
Fonix speech solutions are made for individual end-users as well as for systems. Some of our end-user applications include Fonix VoiceDialTM, Fonix VoiceCentralTM, and Fonix ConnectMeTM.

Fonix VoiceDial is a totally interactive, hands-free software application for Windows Mobile Pocket PC Phone Edition devices and Smartphones, which enables users to place calls by number or by contact. Users simply speak the number or name to call, and VoiceDial does the rest.

Fonix VoiceCentral packages the convenience of VoiceDial with additional voice-enabled capabilities. Users may launch or close applications; listen to email; reply to messages; access calendar, tasks and Internet; and dial contacts with VoiceDial's superior interface.

Fonix ConnectMe is a speech recognition telephone attendant/operator that provides an efficient, professional means of routing incoming, outgoing and internal calls. Customers and employees dial one number, speak the name of the person or department they wish to reach, and are instantly connected.
What SDKs are available to manufacturers and developers?
Fonix offers a complete array of speech technology engines available as SDKs to meet the needs of developers and device manufacturers. We sell Fonix automatic speech recognition (ASR) and text-to-speech (TTS) engines.

With Fonix speech SDKs, developers can quickly and easily integrate robust speech interfaces into embedded products such as:
  • Medical instruments
  • Translators
  • Language learning devices
  • Mobile phones and PDAs
  • Data collection devices
  • Entertainment systems
  • Appliances
  • Games, toys and education devices

Fonix VoiceIn™ ASR
Integrating Fonix VoiceIn into applications and devices allows developers to add command-and-control functions that respond to continuous speech, enabling users to speak naturally to devices. Fonix proprietary neural network technology does not require the user to train the system to recognize his or her speech (a.k.a. speaker independent). Because of the nature of the technology, noisy environments have less impact on Fonix ASR, providing a higher degree of accuracy and a wider array of applications and device usage. Unlike other speech interfaces that are "shrunk-to-fit" embedded devices, Fonix ASR was specifically designed for use limited-memory devices --- "built-to-fit" when developers have memory and MIPs constraints.

FonixTalk™ TTS

FonixTalk is the smallest footprint, full-featured, multi-language TTS engine on the market. No other TTS solution can provide the same flexibility, feature set and voice quality. FonixTalk is the best TTS solution for limited-memory devices or applications, making the user's experience more functional, convenient and enjoyable. FonixTalk adds a high degree of intelligibility to provide the best overall set of TTS voices available. FonixTalk uses a vocal tract model based on a new High Level Synthesis (HL Syn) technology to create a more natural voice -- one that includes inflections, intonations, pauses and changes in pitch, speed and emphasis. The result is speech that's easier to understand. FonixTalk is today's most intelligible TTS technology on the market

What are some other uses of Fonix speech technology?
In addition to handheld mobile/wireless devices, automobiles, PC and web applications, Fonix technology can also be applied to systems.

We provide telephony and server-based solutions for automated phone directory and database information systems. We believe that traditional operator systems and other means of accessing information are becoming antiquated. Significant employee and personal time is lost trying to access information through keypad directories or because calls are blocked after hours. Also, information stored or transferred through servers, PBX or databases may not easily be accessed through non-integrated platforms. Voice-automated systems are capable of integrating these markets and meeting customer expectations of competitive costs and easy installation with minimal change to their existing infrastructure and a simple user interface.

Fonix speech solutions for telephony and server systems include:

Fonix ConnectMeTM
Fonix ConnectMe is a unique voice-automated telephone operator that provides an efficient, professional means of routing incoming, outgoing and internal calls. Customers and employees dial one number, speak the name of a person or department and are quickly connected to the person or department they want to reach. Whether during peak business hours or late at night, ConnectMe's 24-hour high-tech customer service capabilities ensure that all calls reach their intended destinations. By deploying ConnectMe, businesses increase employee efficiency and decrease the amount of time receptionists and office managers spend answering phones. ConnectMe handles all incoming calls simultaneously, so callers are never put on hold. It eliminates the annoyance of remembering and dialing extension numbers or looking up extensions in a phone directory. Employees can create, maintain and access their own phone lists, and can customize the delivery of calls. For example, while at lunch, employees can route calls to their cell phones or pre-define ConnectMe for weekends and holidays by creating a weekday schedule. To our knowledge, no other company in this niche has emerged with a competitive product with our unique features, functionality and price.

511 Traffic Information System
In July 2000, the Federal Communications Commission (FCC) assigned "511" as the number for nationwide access to traveler information. 511 was designated as a free service and when fully implemented will cover the majority of roads in the U.S., helping travelers avoid congested routes and safety hazards. By dialing 5-1-1, callers can access information about route-specific weather and road conditions. Fonix and partner Meridian Environmental Technology, Inc., a leading provider of weather forecasting and analysis services in the Midwest, provide a speech interface 511 system in North Dakota. The system uses Fonix ASR technology to give callers the easiest and safest interface to getting road condition and weather information. The caller just asks for information on a specific section of highway, and the application using Fonix's speech interface queries a database for that highway and speaks the information from the database to the caller. Current markets in development include Nebraska, Montana and South Dakota, with other states scheduled to deploy the system in the near future.
Does Fonix describe its products as having speech or voice recognition?
Earlier in the industry's history, we always referred to products that included automatic speech recognition (ASR). That phrase has morphed over time to include speech recognition (SR) and voice recognition (VR). It's somewhat arbitrary and/or market dependent. The preferred phrase in the Asian market is SR, so that's usually the verbiage used in press releases and marketing materials for partners and customers like Seiko Epson. In the games market, the phrase "voice command and control" is often used, and you'll rarely see the term speech recognition.
Is Fonix ASR technology speaker dependent or speaker independent?
Fonix speech recognition technologies are speaker independent, which is a huge benefit to end-users. There's no speaker training required, and the technology is designed to work in noisy environments.
What languages do Fonix solutions support?
Fonix speech solutions support a variety of languages (variable depending on the specific product), including English and UK English, European French, Canadian French (ASR only), German, Latin American and Castilian Spanish, Italian, Swedish (ASR only), Japanese (ASR only), Korean and Mandarin Chinese (FonixTalk 6.1 only).
What customer support is available?
Our goal at Fonix is to help customers and developers solve problems and answer questions as quickly and easily as possible. To do so, we have a number of resources available for most of our products such as FAQs (Frequently Asked Questions), Product Manuals, White Papers and Service Releases. If you are unable to find answers to questions after searching the product pages of our Website, you may find product-specific contact information for our support team on the "Support" page of each product.
Are Fonix solutions compatible with existing network protocols and data transmission standards?
Fonix speech engines support PCM and G.711 audio encoding and flash-hook transfer for T/R lines. We also support network standards on turnkey products and via partners who develop applications for wired and wireless networks. We plan to remain compatible with transmission protocols and other industry standards as our product line expands.
Are Fonix solutions compatible with new technologies on the horizon?
In the future, there will be a significant increase in the use of speech interfaces in automobiles for hands-free telephony and the control of non-critical systems. This same hands-free speech technology will become universal for mobile phones and PDA / wireless communicators. Games, toys, home appliances and even "smart" home systems (like thermostat, security, multi-media and appliance control) will incorporate speech recognition and voice command and control functionality. The speaker-independent, remote-microphone and noise-robust technologies that Fonix provides today are the market drivers.
What are Fonix’s plans for expansion?
Fonix will continue to couple our award-winning technology with the leading names in mobile phones and PDAs, electronic devices, entertainment games platforms and telephony solutions. We intend to expand awareness of our products and services and enhance our competitiveness by increasing value-added solutions, portability and ease of use.
How can I inquire about employment opportunities at Fonix?
Fonix always looks for skilled, enthusiastic people to contribute to our team and help us provide leading-edge speech solutions to our partners and customers. Qualified applicants interested in sales, engineering, or customer service careers at Fonix Corporation may contact Stacy Hansen in Fonix Human Resources at shansen@fonix.com.