# US 2021/0343271 A1 · Method and apparatus for correcting failures in automated speech recognition systems > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0343271 A1 - Document ID: US20210343271A1 - Status: Published application - Kind: A1 - Application number: 16/866,009 - Filed: 2020-05-04 - Published: 2021-11-04 - Earliest claimed: 2020-05-04 - Assignees: Rovi Guides, Inc. - Applicants: Rovi Guides, Inc. - Inventors: Ankur Anil Aher; Jeffry Copps Robert Jose - Web page: https://patentgrep.com/patent/US20210343271A1 ## CPC - G10L 15/01: Assessment or evaluation of speech recognition systems - G06F 16/90332: {Natural language query formulation or dialogue systems} - G10L 15/063: {Training} - G10L 15/075: {supervised, i.e. under machine guidance} - G10L 15/1815: {Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning} - G10L 15/20: Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech (G10L21/02 takes precedence) - G10L 15/32: Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Systems and methods are disclosed and described for correcting errors in ASR transcriptions. For an incorrect transcription, different words or phrases from the transcription, and/or related words or phrases, are submitted as hint words to the ASR system, and the voice query is submitted again, to determine new transcriptions. This process is repeated with different transcription terms, until a different and more proper transcription is generated. This increases the accuracy of ASR systems. ## Claims ### Claim 1 A method of correcting errors in searches for electronic content resulting from incorrect transcriptions by an automated speech recognition system, the method comprising: for an incorrect transcription output by one or more automated speech recognition systems, successively submitting, using control circuitry, terms associated with different subsets of terms of the transcription as hint words to the one or more automated speech recognition systems, until a resulting transcription output by the one or more automated speech recognition systems differs from the incorrect transcription; and submitting the resulting transcription to an electronic content search system for selecting electronic content corresponding to the resulting transcription. ### Claim 2 (depends on 1) The method of claim 1, further comprising: determining an occurrence of an improper search conducted by the electronic content search system; and determining whether the occurrence of the improper search resulted from the incorrect transcription. ### Claim 3 (depends on 2) The method of claim 2, wherein the determining an occurrence of an improper search further comprises determining the occurrence of an improper search according to one or more of: whether no electronic content was selected by the electronic content search system as a result of the search; whether electronic content selected by the electronic content search system as a result of the search has an associated relevance score less than a predetermined value; whether the terms of the transcription are not connected or weakly connected in a knowledge graph of terms of the electronic content; or whether the terms of the transcription have popularity scores below a predetermined value. ### Claim 4 (depends on 2) The method of claim 2, wherein the determining that the occurrence of the improper search resulted from the incorrect transcription further comprises determining the presence of the incorrect transcription according to one or more of: whether a transcription score of the transcription corresponding to the improper search is less than a predetermined value; whether one or more terms of the transcription corresponding to the improper search have a homonym; whether one or more terms of the transcription corresponding to the improper search have a corresponding pronunciation difficulty metric greater than a predetermined value; a noise level of a query resulting in the transcription corresponding to the improper search; whether one or more terms of the transcription corresponding to the improper search is a mispronounced term; or whether the transcription corresponding to the improper search comprises more than one term. ### Claim 5 (depends on 2) The method of claim 2, further comprising, if the electronic content search system did not generate an occurrence of the improper search, transmitting for display representations of the electronic content selected by the electronic content search system. ### Claim 6 (depends on 2) The method of claim 2, further comprising conditionally performing the successively submitting if the electronic content search system generated an occurrence of the improper search and if the occurrence of the improper search resulted from the incorrect transcription. ### Claim 7 (depends on 2) The method of claim 2, further comprising, if the electronic content search system generated an occurrence of the improper search and if the occurrence of the improper search did not result from the incorrect transcription, generating an error message and transmitting for display representations of the electronic content selected by the electronic content search system. ### Claim 8 (depends on 1) The method of claim 1, further comprising, if none of the different subsets results in a resulting transcription different from the incorrect transcription, generating an error message and generating for display representations of the electronic content selected by the electronic content search system. ### Claim 9 (depends on 1) The method of claim 1, wherein the successively submitting further comprises successively submitting the different subsets as hint words to the one or more automated speech recognition systems, until multiple ones of the resulting transcriptions differ from the incorrect transcription. ### Claim 10 (depends on 9) The method of claim 9, further comprising selecting a most common transcription of the resulting transcriptions that differ from the incorrect transcription, submitting the most common transcription to an electronic content search system for selecting electronic content corresponding to the most common transcription, and transmitting for display representations of the electronic content selected by the electronic content search system from the most common transcription. ### Claim 11 A system for correcting errors in searches for electronic content resulting from incorrect transcriptions by an automated speech recognition system, the system comprising: a storage device; and control circuitry configured to: for an incorrect transcription output by one or more automated speech recognition systems, successively submit terms associated with different subsets of terms of the transcription as hint words to the one or more automated speech recognition systems, until a resulting transcription output by the one or more automated speech recognition systems differs from the incorrect transcription; and submit the resulting transcription to an electronic content search system for selecting electronic content corresponding to the resulting transcription. ### Claim 12 (depends on 11) The system of claim 11, wherein the control circuitry is further configured to: determine an occurrence of an improper search conducted by the electronic content search system; and determine whether the occurrence of the improper search resulted from the incorrect transcription. ### Claim 13 (depends on 12) The system of claim 12, wherein the determining an occurrence of an improper search further comprises determining the occurrence of an improper search according to one or more of: whether no electronic content was selected by the electronic content search system as a result of the search; whether electronic content selected by the electronic content search system as a result of the search has an associated relevance score less than a predetermined value; whether the terms of the transcription are not connected or weakly connected in a knowledge graph of terms of the electronic content; or whether the terms of the transcription have popularity scores below a predetermined value. ### Claim 14 (depends on 12) The system of claim 12, wherein the determining that the occurrence of the improper search resulted from the incorrect transcription further comprises determining the presence of the incorrect transcription according to one or more of: whether a transcription score of the transcription corresponding to the improper search is less than a predetermined value; whether one or more terms of the transcription corresponding to the improper search have a homonym; whether one or more terms of the transcription corresponding to the improper search have a corresponding pronunciation difficulty metric greater than a predetermined value; a noise level of a query resulting in the transcription corresponding to the improper search; whether one or more terms of the transcription corresponding to the improper search is a mispronounced term; or whether the transcription corresponding to the improper search comprises more than one term. ### Claim 15 (depends on 12) The system of claim 12, wherein the control circuitry is further configured to, if the electronic content search system did not generate an occurrence of the improper search, transmit for display representations of the electronic content selected by the electronic content search system. ### Claim 16 (depends on 12) The system of claim 12, wherein the control circuitry is further configured to conditionally perform the successively submitting if the electronic content search system generated an occurrence of the improper search and if the occurrence of the improper search resulted from the incorrect transcription. ### Claim 17 (depends on 12) The system of claim 12, wherein the control circuitry is further configured to, if the electronic content search system generated an occurrence of the improper search and if the occurrence of the improper search did not result from the incorrect transcription, generate an error message and transmitting for display representations of the electronic content selected by the electronic content search system. ### Claim 18 (depends on 11) The system of claim 11, wherein the control circuitry is further configured to, if none of the different subsets results in a resulting transcription different from the incorrect transcription, generate an error message and generating for display representations of the electronic content selected by the electronic content search system. ### Claim 19 (depends on 11) The system of claim 11, wherein the successively submitting further comprises successively submitting the different subsets as hint words to the one or more automated speech recognition systems, until multiple ones of the resulting transcriptions differ from the incorrect transcription. ### Claim 20 (depends on 19) The system of claim 19, wherein the control circuitry is further configured to select a most common transcription of the resulting transcriptions that differ from the incorrect transcription, submit the most common transcription to an electronic content search system for selecting electronic content corresponding to the most common transcription, and transmit for display representations of the electronic content selected by the electronic content search system from the most common transcription. ### Claim at position 21 21-30. (canceled)