Abstract: A speech processing method including the step of identifying prerecorded audio comprising an original speech segment and a corresponding original noise segment. An audio stream can be generated from the prerecorded audio. The audio stream can comprise a stream speech segment and a stream noise segment. The stream speech segment can have approximately a same duration as the original speech segment. The stream noise segment can have a longer duration than the original noise segment. The audio stream can be conveyed to a speech recognition engine. The speech recognition engine can automatically determine an end of utterance condition based upon the stream noise segment. The original noise segment can be of insufficient duration for the speech recognition engine to determine the end of utterance condition. Responsive to the determining of the end of utterance condition, the stream speech segment can be speech recognized.

Abstract: An in-vehicle system that shares speech processing resources among multiple applications located within a vehicle. The system can include one or more software applications, each associated with different functionally independent in-vehicle consoles. Each application can have a console specific user interface. The system can also include a single in-vehicle speech processing system implemented separately from the in-vehicle consoles. The speech processing system can execute speech processing tasks responsive to requests received from the applications. That is, the in-vehicle speech processing system can provide speech processing capabilities for the applications. The provided speech processing capabilities can include text-to-speech capabilities and speech recognition capabilities.

Abstract: The present invention records a loudness-level-reference segment of audio when creating speech audio files and audio files including background sounds. The speech audio files can then be combined with the background sound containing audio files in any desirable combination. When combining the files, the relative audio level of the files is matched, by matching the loudness-level-reference segments with each other. Any of a variety of known digital signal processing techniques can be used to normalize the component audio files. The combined audio files containing speech and background sounds (e.g. ambient noise) having matching relative audio levels can be used to test and/or train a speech recognition engine or a speech processing system.

Abstract: The present invention records a loudness-level-reference segment of audio when creating speech audio files and audio files including background sounds. The speech audio files can then be combined with the background sound containing audio files in any desirable combination. When combining the files, the relative audio level of the files is matched, by matching the loudness-level-reference segments with each other. Any of a variety of known digital signal processing techniques can be used to normalize the component audio files. The combined audio files containing speech and background sounds (e.g. ambient noise) having matching relative audio levels can be used to test and/or train a speech recognition engine or a speech processing system.