| 摘要 |
Compression of voice and data signals by at least an order of magnitude that permits greater use of the limited bandwidth at the low frequency of operation. A voice recognition system that converts words spoken into a microphone into a sequence of letters and gaps. An encoder codes the letters into a digital message. A transmitter transmits the digital message to a receiver over a communications link. The received digital message is decoded in a decoder and a speech synthesizer converts the decoded message into spoken words that are annunciated on a speaker. In addition, an initial message that identifies a stored voice type that is to be synthesized, or a simultaneous signal that tailors the voice synthesizer in real time as the voice of the speaker changes may be transmitted to cause the speech synthesizer to more accurately resemble a speaker's voice. A speech-to-text processor and a display may be used to display the message at the receiver for hearing impaired users and users located in noisy areas. This feature may also be used to transmit messages to a digital storage device coupled to the speech-to-text processor. A digital interface to the transmitter may be provided for nonspeaking individuals to communicate using a keypad or similar device to generate the message to be transmitted. This feature may also be used for transmitting stored files to the receiver.
|
