Method for performing speech recognition and processing system
A method for performing speech recognition relating to an object for the purpose of affecting automatic processing of the object by a processing system. The object carries information with at least a character string of processing information. The character string spoken by an operator is processed by way of a speech recognition procedure to generate a first result. Based on the need for more information of an element of the first result additional processing data is requested. An operator's response generates a second result. The first result is then modified to achieve consistency with the operator's response.
1. A method of performing speech recognition on an object for automatic processing of the object by a processing system, wherein the object includes an information area containing a character string of processing information, the method comprising:
acquiring a signal representing a sequence of characters spoken by an operator corresponding to the character string and processing the signal with a speech recognition processor by way of a speech recognition procedure generating a first ambiguous speech recognition result;
based on specific characteristics of the first ambiguous speech recognition result, requesting additional processing information from the operator by providing request information with the processing system;
acquiring a signal representing an operator's response to the request and processing the operator's response with the speech recognition processor for generating a second ambiguous speech recognition result;
modifying the first ambiguous speech recognition result to achieve consistency with the operator's response; and
making a digital image of the information area available for an optical character recognition (OCR) procedure and performing the OCR procedure on the digital image using at least a part of the modified ambiguous speech recognition result for the OCR procedure.
2. The method according to claim 1 , wherein the first ambiguous speech recognition result is a candidate list with a plurality of candidates, at least one of the plurality of candidates corresponding to the character string.
3. The method according to claim 1 , wherein a specific characteristics for requesting additional processing information is that at least a part of the first ambiguous speech recognition result is stored in a data memory of the processing system as data needing additional processing information.
4. The method according to claim 1 , which comprises deriving the request information from the first ambiguous speech recognition result, the request information being different from the character string.
5. The method according to claim 1 , wherein the request information contains processing information.
6. The method according to claim 1 , which comprises processing the operator's response to generated a restricted vocabulary for speech recognition and generating the second ambiguous speech recognition result from the restricted vocabulary.
7. The method according to claim 1 , wherein the operator's response is a further character string of the processing information of the object.
8. The method according to claim 1 , wherein the character string is a ZIP code and the request information is a city name.
9. The method according to claim 8 , wherein the request information corresponds to an element of the first ambiguous speech recognition result, the element being inconsistent with the character string.
10. The method according to claim 1 , wherein the operator's response is a negation if the request information is not consistent with the processing information of the object.
11. The method according to claim 1 , wherein the modifying step comprises removing an element of the first ambiguous speech recognition result.
12. The method according to claim 1 , wherein, if the operator's response is a negation, the modifying step comprises removing from the first ambiguous speech recognition result a portion corresponding to the request information.
13. The method according to claim 1 , wherein the modifying step comprises removing from the first ambiguous speech recognition result an element not consistent with the second ambiguous speech recognition result.
14. The method according to claim 1 , which comprises:
providing the request information to the operator as verification request for verifying the request information by giving verification information;
processing the operator's response upon the request as processing information to generate a second ambiguous speech recognition result; and
processing the verified request information as correct processing information.
15. A processing system for the automatic processing of an object, the object having an information area containing at least a character string of processing information, comprising:
a speech recognition system having a port configured to be coupled to a communication device of an operator for inputting at least a sequence of characters, said speech recognition system being configured to generate a first ambiguous speech recognition result from an input received from the communication device;
a controller connected to said speech recognition system, said controller being configured:
based on specific characteristics of the first ambiguous speech recognition result, to control a request for additional processing data by providing request information to the operator;
to control a processing of an operator's response to the request to generate a second ambiguous speech recognition result; and
to modify the first ambiguous speech recognition result to achieve consistency with the second ambiguous speech recognition result; and
an optical character recognition (OCR) system connected to said controller and configured to recognize characters of the character string on the object using at least a part of the modified ambiguous speech recognition result.