Source-based automatic speech recognition

Patent metadata

Field	Value
Publication number	US-10950239-B2
Application number	US-201514920021-A
Country	US
Kind code	B2
Filing date	Oct 22, 2015
Priority date	Oct 22, 2015
Publication date	Mar 16, 2021
Grant date	Mar 16, 2021

Abstract

Recognizing a user's speech is a computationally demanding task. If a user calls a destination server, little may be known about the user or the user's speech profile. The user's source system (device and/or server) may have an extensive profile of the user. As provided herein, a source device may provide translated text and/or speech attributes to a destination server. As a benefit, the recognition algorithm may be well tuned to the user and provide the recognized content to the destination. Additionally, the destination may provide domain attributes to allow the source recognition engine to better recognize the spoken content.

First claim

What is claimed is:

1. A destination server, comprising:
a network interface to a communications network;
a microprocessor having access to the network interface; and
the microprocessor that, via the network interface, engages in a call with a source server, the call comprising a voice channel comprising a spoken portion provided by a source user and a data channel comprising a machine-readable cue of the spoken portion and wherein the machine-readable cue comprises a speech attribute of the source user;
wherein the microprocessor executes a speech recognition algorithm to recognize the spoken portion and wherein the speech recognition algorithm is seeded with the machine-readable cue; and
wherein the microprocessor executes instructions in accordance with the microprocessor recognized spoken portion on the voice channel and the machine-readable cue received on the data channel.

2. The destination server of claim 1 , wherein the microprocessor receives indicia of source-side speech recognition.

3. The destination server of claim 2 , wherein the microprocessor, in response to receiving the indicia of source-side speech recognition, replies via the data channel with a domain attribute associated with the destination server.

4. The destination server of claim 1 , wherein the microprocessor executes a speech recognition algorithm utilizing an acoustic model, selected in accordance with the speech attribute of the source user in the machine-readable cue, and derives machine-readable content from a waveform portion of the call.

5. The destination server of claim 1 , wherein the machine-readable cue further comprises human-readable text of a machine-readable recognition of the spoken portion.

6. The destination server of claim 1 , wherein the data channel comprises a Real-Time Transport Protocol (RTP) text stream.

Assignees

• Avaya Inc

Inventors

• Lin Lin
• Lin Ping

Classifications

• G10L2015/226

  using non-speech characteristics · CPC title

• G10L15/32

  Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems · CPC title

• G10L15/30 Primary

  Distributed recognition, e.g. in client-server systems, for mobile phones or network applications · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06F40/58

  Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation · CPC title