Patent Application
20080162129
This invention relates generally to speech recognition processes and more particularly to speech recognition search processes.
Speech recognition comprises a known area of endeavor. Certain speech recognition processes make use of speech recognition search processes such as, but not limited to, the so-called hidden Markov model-based speech recognition process. This generally comprises use of a statistical model that outputs a sequence of symbols or quantities where speech is essentially treated as a Markov model for stochastic processes commonly referred to as states. An exemplary hidden Markov model might output, for example, a sequence of 39-dimensional real-valued vectors, outputting one of these about every 10 milliseconds.
Such vectors might comprise, for example, cepstral coefficients that are obtained by taking a Fourier transform of a short-time window of sampled speech and de-correlating the spectrum using a cosine transform, then taking the first (most significant) coefficients for these purposes. The hidden Markov model approach will tend to have, for each state, a statistical distribution called a mixture of diagonal or full covariance Gaussians that will characterize a corresponding likelihood for each observed vector.
In many prior art approaches, a conventional speech recognition search requires that boundaries between words, subwords, and the aforementioned states be searched on a regular basis (typically per each frame of sampled audio content) using a single level of resolution. Though indeed an optimal and powerful approach, this frame-by-frame (or single resolution) approach to searching for word, subword, and state boundaries also requires considerable computational resources. This need only grows with the depth and richness of the supported vocabulary. As a result, a speech recognition process that employs a speech recognition search process can require enormous computational resources.
Consider, for example, an application setting where each frame represents only about 10 milliseconds of audio content. For a speech recognition process that supports recognition of, say, 50,000 words, it then becomes necessary to search and compare the recognition data as corresponds to each of those 50,000 words for each such frame. This, alone, can require considerable computational capability. These requirements only grow more severe as one considers that such a process also requires a corresponding search for subwords with each such frame.
As a result, such an approach, while often successful to carry out optimal speech recognition, is also often too computationally needy to work well in an application setting where such computational overhead is simply not available. Small, portable, wireless communications devices such as cellular telephones and the like, for example, represent such an application setting. Both available computational capability as well as corresponding power capacity limitations can severely limit the practical usage of such an approach.
The above needs are at least partially met through provision of the method and apparatus pertaining to the processing of sampled audio content using a speech recognition search process described in the following detailed description, particularly when studied in conjunction with the drawings, wherein:
Skilled artisans will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions and/or relative positioning of some of the elements in the figures may be exaggerated relative to other elements to help to improve understanding of various embodiments of the present invention. Also, common but well-understood elements that are useful or necessary in a commercially feasible embodiment are often not depicted in order to facilitate a less obstructed view of these various embodiments of the present invention. It will further be appreciated that certain actions and/or steps may be described or depicted in a particular order of occurrence while those skilled in the art will understand that such specificity with respect to sequence is not actually required. It will also be understood that the terms and expressions used herein have the ordinary meaning as is accorded to such terms and expressions with respect to their corresponding respective areas of inquiry and study except where specific meanings have otherwise been set forth herein.
Generally speaking, pursuant to these various embodiments, one provides a plurality of frames of sampled audio content and then processes that plurality of frames using a speech recognition search process that comprises, at least in part, searching for at least one of state boundaries at a base resolution, for example, within each frame, searching for at least two of state boundaries, subword boundaries, and word boundaries using different search resolutions. This contrasts sharply with present practice, of course, in that present practice will typically require systematically searching each frame (or at single resolution) for each state, subword, and word boundaries.
This can comprise, by one approach, using a first relatively fine level of search resolution (such as each and every frame) when searching for state boundaries and a coarser level of resolution (such as every other frame) when searching for sub-word and word boundaries. As another example, this can comprise, by one approach, using a first relatively fine level of search resolution (such as each and every frame) when searching for state boundaries, a coarser level of resolution (such as every other frame) when searching for sub-word boundaries, and an even courser level of resolution (such as every fourth frame) when searching for word boundaries.
So configured, these teachings permit relatively accurate and high quality speech recognition processing as one might ordinarily expect when using such speech recognition search processes while nevertheless avoiding a considerable amount of computational activity. By skipping some frames in this regard, the processing platform can be significantly relieved of the corresponding computational support. This, in turn, permits a given processing platform having only modest capacity and/or capability to nevertheless often carry out a speech recognition search process with successful results.
These and other benefits may become clearer upon making a thorough review and study of the following detailed description. Referring now to the drawings, and in particular to
The above-mentioned speech recognition search process can comprise such processes as may be suitable to meet the needs of a given application setting. For the purposes of providing an illustrative example and not by way of limitation it will be presumed herein that this speech recognition search process comprises a hidden Markov model-based speech recognition process.
By one approach, this step 102 can comprise searching for each of state boundaries, subword boundaries, and word boundaries using a base resolution, secondary resolution, and third resolution, respectively that are each different from one another. This can comprise, for example, searching for state boundaries for every frame, only searching for subword boundaries for every Nth frame (where N comprises an integer larger than one) and only searching for word boundaries for every Mth frame (where M comprises an integer equal to or larger than N and, more particularly, may comprise an integer that comprises a multiple of N).
To illustrate, consider the schematic representation shown in
So configured, those skilled in the art will recognize and appreciate that the overhead requirements associated with subword boundary searching is halved and the overhead requirements associated with word boundary searching is reduced by 75%. This, of course, represents a considerable reduction in computational requirements and makes such a speech recognition search process available to a greatly increased population of platforms including, for example, cellular telephones and the like.
Those skilled in the art will recognize that greater savings in this regard are achieved by increasing the number of skipped frames. Such an increase, however, at some point may reduce the overall quality of the speech recognition process. The appropriate settings to apply in a given situation may change with the application setting as the designer strikes a satisfactory compromise between the quality of the resultant output and corresponding computational requirements.
Those skilled in the art will appreciate that the above-described processes are readily enabled using any of a wide variety of available and/or readily configured platforms, including partially or wholly programmable platforms as are known in the art or dedicated purpose platforms as may be desired for some applications. Referring now to
In this example, the implementing apparatus 300 comprises an input 302 that operably couples to a processor 301. The input 302 can be configured and arranged to provide a plurality of frames of sampled audio content. Again, there are various known ways by which this can be accomplished that will be readily known and available to a person skilled in the art. The processor 301, in turn, can comprise a dedicated purpose or a partially or wholly programmable platform that is configured and arranged (via, for example, corresponding programming) to effect selected teachings as have been set forth herein. In particular, this processor 301 can be configured and arranged to process the incoming plurality of frames using a speech recognition search process that comprises, at least in part, the aforementioned searching for at least one of subword boundaries and word boundaries as may be contained within each frame less often than on a frame-by-frame basis.
Those skilled in the art will recognize and understand that such an apparatus 300 may be comprised of a plurality of physically distinct elements as is suggested by the illustration shown in
So configured, an implementing platform having only modest processing capabilities (such as a cellular telephone or the like) can nevertheless make highly leveraged use of powerful speech recognition search processes by effectively skipping some frames on a regular basis when searching for subword and/or word boundaries as may be contained within such frames. The described approaches are relatively easy to implement and are also readily scaled to meet the needs and/or opportunities as correspond to a given application setting. For
Those skilled in the art will recognize that a wide variety of modifications, alterations, and combinations can be made with respect to the above described embodiments without departing from the spirit and scope of the invention, and that such modifications, alterations, and combinations are to be viewed as being within the ambit of the inventive concept.
This application is related to a U.S. application being filed on the same date, having attorney docket number CML040301HI, entitled METHOD AND APPARATUS PERTAINING TO THE PROCESSING OF SAMPLED AUDIO CONTENT USING A FAST SPEECH RECOGNITION SEARCH PROCESS, having inventor Yan Ming Cheng, and assigned to the assignee hereof. The USASN of the related application is unknown at this time.
