Streamlined navigational speech recognition

• US 10,008,201 B2
• Filed: 09/28/2015
• Issued: 06/26/2018
• Est. Priority Date: 09/28/2015
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Talwar, Gaurav, Hecht, Ron M., Zhao, Xufang
• Primary Examiner(s): Sharma, Neeraj
• Application Number:
  US14/867,574
• Publication Number:
  US 20170092295A1
• Time in Patent Office: 1,002 Days
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules

Abstract

A system and method of performing automatic speech recognition (ASR) includes: receiving speech at a vehicle microphone; communicating the received speech to an ASR system; measuring an amount of time that elapses while speech is received; selecting a point-of-interest (POI) context or an address context based on the measured amount of received time; and processing the received speech using a POI context-based grammar when a POI context is selected or an address-based grammar when an address context is selected.

Claims

1. A method of performing automatic speech recognition (ASR), comprising:

(a) establishing a point-of-interest (POI) time window for a received spoken request;

(b) establishing an address time window for the received spoken request;

(c) receiving the spoken request at a vehicle microphone;

(d) measuring an amount of time that elapses while the spoken request is received;

(e) determining that the amount of time fails within both the POI time window and the address time window;

(f) performing a keyword search of the received spoken request in response to step (e); and

(g) determining a POI context or an address context based on the keyword search without recognizing a gateway command.

2. The method of claim 1, wherein the POI time window and the address time window overlap.

3. The method of claim 2, further comprising the step of performing a keyword search using a grammar optimized for a POI and an address context.

4. The method of claim 1, wherein the ASR system is remotely located from a vehicle.

5. A method of performing automatic speech recognition (ASR), comprising:

(a) establishing an address time window for a received spoken request;

(b) receiving the spoken request at a vehicle microphone;

(c) communicating the received spoken request to an ASR system;

(d) measuring an amount of time that elapses while the spoken request is received;

(e) detecting the presence of one or more digit strings in the received spoken request received during the first half of the measured amount of time;

(f) determining that at least some of the received spoken request is received during the address time window; and

(g) processing the received spoken request using an address-based grammar in response to step (e) and step (f) without recognizing a gateway command.

6. The method of claim 5, wherein the ASR system is remotely located from a vehicle.