One of the most natural of all data entry technologies is the human voice. No pre-printed symbols or marks are required, no encoded tags are needed. Operators speak into a microphone and the words are translated into computer input.

Although there are very large systems currently used in telephony systems that recognize thousands of speakers using very large vocabularies, these systems tend to require the power of a mainframe. In data collection, speech recognition systems are run on small, dedicated devices or are integrated into hand-held terminals which do not have such immense computing horsepower.

Available speech recognition systems for industrial use range from:

• Relatively limited with a small number of words,
• Complex ones that can recognize hundreds, even thousands, of words,
• Systems that recognize all speakers (limited words), and
• Systems that must be trained (extensive vocabularies).

Speech recognition systems are currently being used for:

• Inspection reporting,
• Quality control,
• Applications where words are the desired input, and
• Eyes- and hands-free operation is required.

Speech Recognition Advantages:

• A "natural" data input methodology,
• Eyes- and hands-free operation,
• High security (typically must be "trained" for each user),
• Reliability,
• Flexibility (language)

Speech Recognition Limitations:

• Not optimal for entry of strings of numbers or letters.