# US 2013/0159000 A1 · Spoken Utterance Classification Training for a Speech Recognition System > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2013/0159000 A1 - Document ID: US20130159000A1 - Status: Published application - Kind: A1 - Application number: 13/326,659 - Filed: 2011-12-15 - Published: 2013-06-20 - Earliest claimed: 2011-12-15 - Assignees: Microsoft Corporation - Inventors: Yun-Cheng Ju; James Garnet Droppo, III - Web page: https://patentgrep.com/patent/US20130159000A1 ## CPC - G10L 15/1822: {Parsing for meaning understanding} - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract The subject disclosure is directed towards training a classifier for spoken utterances without relying on human-assistance. The spoken utterances may be related to a voice menu program for which a speech comprehension component interprets the spoken utterances into voice menu options. The speech comprehension component provides confirmations to some of the spoken utterances in order to accurately assign a semantic label. For each spoken utterance with a denied confirmation, the speech comprehension component automatically generates a pseudo-semantic label that is consistent with the denied confirmation and selected from a set of potential semantic labels and updates a classification model associated with the classifier using the pseudo-semantic label. ## Claims ### Claim 1 In a computing environment, a method performed at least in part on at least one processor, comprising, providing a classifier for a speech recognition system, including, processing spoken utterances associated with the speech recognition system, for each spoken utterance having a denied confirmation, assigning a pseudo-semantic label that is consistent with the denied confirmation and selected from a set of potential semantic labels, and updating a classification model associated with the classifier using each assigned pseudo-semantic label. ### Claim 2 (depends on 1) The method of claim 1, wherein assigning the pseudo-semantic label further comprises partitioning the spoken utterances into groupings of spoken utterances having unsettled denied confirmations and a current training set comprising settled spoken utterances, wherein the settled spoken utterances comprise accepted confirmations. ### Claim 3 (depends on 2) The method of claim 2, wherein updating the classification model further comprises for each grouping, determining the pseudo-semantic labels for the spoken utterances having the unsettled denied confirmations and training the classification model with a combination of the current training set and the spoken utterances having the unsettled denied confirmations. ### Claim 4 (depends on 2) The method of claim 2, wherein updating the classification model further comprises for each grouping, determining a number of discrepancies between a semantic labeling produced by the classifier and the combination of the current training set and the spoken utterances having the unsettled denied confirmations. ### Claim 5 (depends on 4) The method of claim 4 further comprising selecting the grouping having a smallest number of discrepancies and incorporating the spoken utterances having the unsettled denied confirmations into the current training set. ### Claim 6 (depends on 5) The method of claim 5 further comprising training the classification model with the current training set and producing classifications results for the spoken utterances having the denied confirmations. ### Claim 7 (depends on 6) The method of claim 6 further comprising updating the pseudo-semantic labels based on the classification results and selecting another grouping having a smallest number of discrepancies to be incorporated into the current training set. ### Claim 8 (depends on 1) The method of claim 1, wherein updating the classification model further comprises training the classification model with a current training set comprising spoken utterances having accepted confirmations and spoken utterances having settled denied confirmations. ### Claim 9 (depends on 1) The method of claim 1 further comprising performing the step of assigning the pseudo-semantic label and the step of updating the classification model for another training iteration. ### Claim 10 (depends on 1) The method of claim 1 further comprising repeating the method if a pseudo-semantic label assignment is modified. ### Claim 11 In a computing environment, a system, comprising, a speech comprehension component configured to interpret spoken utterances using a classifier, wherein the speech comprehension component is further configured to generate output data comprising semantic labels for the spoken utterances and pseudo semantic labels for a portion of the spoken utterances having denied confirmations, to train a classification model associated with the classifier with the pseudo semantic labels and settled ones of the semantic labels and to use the classification model to navigate a speaking entity to desired information. ### Claim 12 (depends on 11) The system of claim 11, wherein the speech comprehension component is further configured to automatically generate a pseudo-semantic label for each spoken utterance having a denied confirmation, wherein the pseudo-semantic label is be consistent with the denied confirmation and satisfies the classification model. ### Claim 13 (depends on 11) The system of claim 11, wherein the speech comprehension component is further configured to identify groupings of spoken utterances having unsettled denied confirmations and to generate current training data comprising spoken utterances having accepted confirmations and spoken utterances having settled denied confirmations. ### Claim 14 (depends on 13) The system of claim 13, wherein the speech comprehension component is further configured to determine pseudo-semantic labels for each grouping based on corresponding ones of the semantic labels and to train the classification model with a combination of the current training data and the each grouping. ### Claim 15 (depends on 14) The system of claim 14, wherein the speech comprehension component is further configured to determine a number of discrepancies that is between updated semantic labels produced by the classifier and the combination of the current training data and one of the groupings of the spoken utterances having unsettled denied confirmations. ### Claim 16 (depends on 15) The system of claim 15, wherein the speech comprehension component is further configured to incorporate the grouping having a smallest number of discrepancies into the current training data, to train the classifier with the current training data and to classify the output data using to produce another set of semantic labels for the spoken utterances. ### Claim 17 (depends on 16) The system of claim 16, wherein the speech comprehension component is further configured to select a remaining one of the grouping to be incorporated into the current training data based on discrepancy indicia. ### Claim 18 One or more computer-readable media having computer-executable instructions, which when executed perform steps, comprising: providing confirmations to callers based on text strings representing spoken utterances in order to route voice calls to appropriate voice menu options; accessing output data comprising a semantic label assignment for the spoken utterances into spoken utterances; assigning pseudo-semantic labels to a portion of the spoken utterances having denied confirmations; and training a classification model with the pseudo-semantic labels and settled ones of the semantic labels. ### Claim 19 (depends on 18) The one or more computer-readable media of claim 18 having further computer-executable instructions comprising: providing a confirmation when the classifier rejected one of the spoken utterance. ### Claim 20 (depends on 18) The one or more computer-readable media of claim 18 having further computer-executable instructions comprising: ignoring any of the spoken utterances that is assigned a reject semantic label.