BACKGROUND OF THE INVENTION In linon digital audio systems, error concealment and audio video (AV) synchronization can require splicing discontinuous audio sample sequences. An error in transmission of a series of audio sample sequences can cause a loss of an audio sample sequence. Known methods and apparatus to compensate for the loss of an audio sample sequence include copying an adjacent audio sample sequence (e. g.. the precedin audio sample sequence. the subsequent audio sample sequence, etc.). Such methods and apparatus can result in a discontinuous audio stream that can cause pops and clicks unless the discontinuity is smoothed.

FIG. 1 shows a known series of audio sample sequences, e. g. first block 101. second block 102. third block 103. Each audio sample sequence can comprise 100 milliseconds (msecs.) of sampled audio data having a varying audio value (e. g., decibels (dBs). etc.) over a period of time. The audio value is continuous from first block 101 to second block 102 to third block 103. FIG. 2 illustrates a known method of compensating for a loss of an audio sample sequence. When the lost audio sequence is the second block 102. then first block 101 can be copied and substituted for the lost second block 10'_'such that a copied first block 202 is positioned between first block 101 and third block 103. Such copying and substituting disadvaniageously can result in discontinuities in the audio stream. e. g.. a discontinuity in the audio stream from first block 101 to copied first block 202. a discontinuity from copied first block 202 to third block 10'.

FIG. 3 shows an illustration of a known method of joinin F discontinuous audio sample sequences. The known method includes ramping down a preceding sequence to zero and ramping up the subsequent sequence from zero. FIG. 3 shows the trailing edge of first block 301 being ramped down to zero. and the leading edge of copied first block 302 being ramped up from zero. Likewise. the trailing edge of copied first block 302 is ramped down to zero. and the leading edge of third block 303 is ramped up from zero.

This method has disadvantages including 1) requiring the need to ramp to zero to be identified soon enough to ramp down the preceding sequence to zero. and 2) the ramp

down to zero and back up from zero is not a minimum achievable distortion of the spliced stream of audio sample sequences.

In view of the foregoing, it can be appreciated that a substantial need exists for embodiments of methods and apparatus that can advantageously smooth spliced discontinuous audio streams.

SUMMARY OF THE INVENTION In one embodiment. a plurality of transition audio sample values can be generated based on a captured value of a final sample of a first audio sample sequence and a received plurality of samples of a second audio sample sequence. Each transition audio sample value can be generated based at least in part on a respective sample of the second audio sample sequence. Each transition audio sample value can represent the value of the captured final sample adjusted with a varying weighted difference between the value of the final sample and a value of the respective sample of the second audio sample sequence. The varying weighted difference can be based on a varying weighting function that increases from a value of zero to a value of one such that the first transition audio sample value is generated based on the varying weighting function having a value of zero. and each subsequent transition audio sample value is generated based on the varying weighting function having a value incrementally greater until the final transition audio sample value is generated based on the varying weighting function having a value equal to one.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows a known series of audio sample sequences.

FIG. 2 illustrates a known method of compensating for a loss of an audio sample sequence.

FIG. 3 shows an illustration of a knows method of joining discontinuous audio sample sequences.

FIG. 4 is an illustration of a method in accordance with an embodiment of the present invention to smooth a spliced discontinuous audio stream.

FIG. shows an apparatus in accordance with an embodiment of the present invention.

FIG. 6 shows an apparatus in accordance with another embodiment of the present invention.

FIG. 7 illustrates a method in accordance with another embodiment of the present invention.

FIG. 8 shows an apparatus in accordance with another embodiment of the present invention.

FIG. 9 shows an apparatus in accordance with another embodiment of the present invention.

DETAILED DESCRIPTION Embodiments of a method and apparatus to smooth a discontinuity of a spliced audio stream are described. In the following description, for purposes of explanation. numerous specific details are set forth to provide a thorough understanding of the present invention. It will be appreciated. however. by one skilled in the art that the present invention may be practiced without these specific details. In other instances. structures and devices are shown in block diagram form. Furthermore. one skilled in the art can readily appreciate that the specific sequences in which methods are presented and performed are illustrative and it is contemplated that the sequences can be varied and stil ! remain within the spirit and scope of the present invention.

FIG. 4 is an illustration of a method in accordance with an embodiment of the present invention to smooth a spliced discontinuous audio stream. First block 403 and copied first block 402 can comprise a series of audio sample sequences of an audio stream that have a time-varying value ofs (t). Each of first block 403 and copied first block 402 can include 10 ( milliseconds of audio samples. The audio value of the audio samples of first block 403 can vary such that the audio value at t = 0 is approximately 40 dBs and the audio value at t = 100 is approximately 60 dBs. First block 403 and copied first block 402 can be joined at a juncture (e. g.. at t = 100 msecs.) such that there is a discontinuity in the value of s (t) around t = 100. At the juncture (e. g.. around t = 100 msecs.). the value of s (t) can drop from approximately 60 dBs to 40 dBs due to the discontinuitv in joining first block 403 to copied first block 40'.

To smooth the discontinuity between first block 403 and copied first blocl ; 40''. aftr.- the junction a transition output audio value can be based on interpolating between s (t) and a delayed sequence of audio samples, s (t-x). In one embodiment. an interpolation method is implemented that replaces a sample stream s (t) with an output based on s (t-x)-f (t) * [s (t)- s (t-x)]. where f (t) is a function that smoothly increases from zero to one. s (t-x) can represent a delayed version of sample stream s (t). where x is a period of time (e. n. a period oí milliseconds. a number of sample periods. etc.).

As illustrated in FIG. 4. s (t) has a value of approximately 60 dBs at t = 100.

Immediatelv after t = 100. s (t) has a value of approximately 40 dBs. Up to and including 100. the output signal can have a value based on s (t). During a juncture transition period. e. u a four millisecond period after t = 100. the value of the output can be based on s (t-x) + f (t) [s (t)-s (t-x)]. where f (t) is a function that smoothly increases from zero to one. and x = four milliseconds. Line 410 shows a delayed sample stream signal s (t-4) from after t = 100 to t = 104. Plotted lined 411 shows the time-varying value of s (t-x) + f (t) * [s (t)-s (t-x)]. where f (t)

is a function that smoothly increases from zero to one (e.g., f(100) = 0, f(101) = .25, f(102) = . 50. f (103) =. 75. f (104) = 1. 0. etc.). and x = four milliseconds.

FIG. 5 shows an apparatus in accordance with an embodiment of the present invention. A multiplexer 450 can select between standard output S 451 and transition output T 452 to provide output o (t). Sample stream s (t). in one embodiment. is communicated tu standard output S 451. s (t) can be communicated to delay 402 to output a delayed sampls stream s (t-x). where x can be a period of time four milliseconds. one or more sample periods. etc.). Subtractor 403 can output the difference between s (t) and s (t-x). e.g., s(t) - s(t- x). Multiplier 404 can output the product of f (t). which is a function that can smoothy increase from zero to one over a period of time. and the output of subtractor 403. Adder 405 can output the sum of the outputs of multiplier 404 and delay 402 such that a transition output signal s (t-x)-f (t) * [s (t)-s (t-x)] is communicated to transition output T 452.

During typical operations (e. .. wllen no copied audio sample sequences are spliced : into the sample stream to replace a lost audio sample sequence. etc.). multiplexer 450 can select the sample stream s (t) present at the standard output S 451 as the output o (t). When n copied audio sample sequence is spliced into the sample stream, the multiplexor 450 can select the transition output T 45 during a transition period to smooth a discontinuity between the spliced audio sample sequences. Subsequent to the transition period. the multiplexor 451 can select the sample stream s (t) present at the standard output S 451 as the output o (t).

For example, referring to FIGS. 4 and 5, multiplexor 450 can output s (t) as output o (t) during the period 0 < t < 100 by outputting the signal present at standard output S 451.

After t = 100, multiplexor 450 can output the transition output signal s (t-x) + f (t) * [s (t)- s (t-x)] as the output o (t) during the period 100 < t < 104 by outputting the signal present at transition output T 452. After z = 104. multiplexer 450 can again output s (t) as the output. o (t) by outputimg the signal present at standard output S 45 i.

FIG. 6 shows an apparatus in accordance with another embodiment of the present invention. A multiplexer 550 can select between standard output S 551. first transition output Tl 552. and second transition output T2 553 to provide output o (t). In accordance with the embodiment illustrated by FIG. 6. a function f (t) can be a function with a dynamic range between zero and one-half. In one embodiment. f (t) smoothly increases from zero to one-half. and then smoothly decreases from one-half to zero.

Sample stream s (t). in one embodiment. is communicated to standard output S 55'. s (t) can be communicated to delay 502 to output a delayed sample stream s (t-x). s-here x ca'. be a period of time four milliseconds. one or more sample periods. etc.). Subtractor 5û 5 can output the difference between s (t) and s (t-x). e. c., s (t)-s (t-x). Multiplier 504 can output the product of f (t). which is a function that can increase from zero to one-half and then decrease from one-half to zero over a period of time. and the output of subtractor 503. Adder

505 can output the sum of the outputs of multiplier 504 and delay 502 such that a first transition output signal s (t-x) + f (t) * [s (t)-s (t-x)] is communicated to first transition output T1 559.

Subtractor DlD can output the difference between s (t-x) and s (t). s (t-x)-s (t).

Multiplier 14 can output the product of f (t). which is a function that can increase from zero to one-half and then decrease from one-half to zero over a period of'time. and the output oi subtractor 513. Adder 515 can output the sum ofs (t) and the output of multiplier 514 sucs that a second transition output signal s (t)-f (t) * [s (t-x,-s (t) l is communicated to second transition output T2 553.

During typical operations when no copied audio sample sequences are spliced into the sample stream to replace a lost audio sample sequence. etc.). multiplexer 550 can select the sample stream s (t) present at the standard output S 5-- as the output o (t). hell. i copied audio sample sequence is spliced into the sample stream, the multiplexor 550 can select the first transition output signal s (t-x) - f(t) * [s(t) - s(t-x)] present at first transition output Tl as the output o (t) during the first half of a transition period. The first half of the transition period can coincide with the period during which f (t) smoothly increases from zero to one-half. After the first half of the transition period, the multiplexor 550 can select the second transition output signal s (t) + f (t) * [s (t-x)-s (t)] present at the second transition output T2 553 as the output o (t) during the second half of the transition period. The second half of the transition period can coincide with the period during which f (t) smoothly decreases from one-half to zero. Subsequent to the transition period, the multiplexor 550 can select the sample stream s (t) present at the standard output S 551 as the output o (t).

FIG. 7 illustrates a method in accordance with another embodiment of the present invention. An audio stream can include a series of audio sample sequences. and a first block 703 and copied first block 702 can be spliced together. Each of first block 703 and copied first block 702 can be an audio sample sequence. A discontinuity in an audio value s (t) can exist at the juncture between first block 703 and copied first block 702. To smooth the discontinuity in s (t). after the junction a plurality of transition audio sample values can be based on interpolating between s (t) and the value ofs (t) at the juncture (i.e., s(tjuncture).

In one embodiment. a plurality of transition audio sample values can be generated based on a captured value of a final sample of a first audio sample sequence and a received plurality of samples of a second audio sample sequence. Each transition audio sample valu_ can be generated based at least in part on a respective sample of the second audio sample sequence. Each transition audio sample value can represent the value of the captured final sample adjusted with a varying weighted difference between the value of the final sample and a value of the respective sample of the second audio sample sequence. The varying weighted difference can be based on a varying weighting function that increases from a value of zero to a value of one such that the first transition audio sample value is generated based on the

varying weighting function having a value of zero. and each subsequent transition audio sample value is generated based on the varying weighting function having a value incrementally greater until the final transition audio sample value is generated based on the varying weighting function having a value equal to one.

In another embodiment. an interpolation method is implemented that replaces a sample stream s (t) with an output based on s(tjuncture) - f(t) * [s(t) - s(tjuncture)]. where f (t) l, a function that smoothly increases from zero to one. As illustrated in FIG. 7. s (t) has a value of approximately 60 dBs at t = 100. Immediately after t = 100. s (t) has a value of approximately 40 dBs. L'p to and including t = 100. the output signal can have a value based on s (t). Durinr a juncture transition period. e. g.. a four millisecond period after t = 100. the value of the output can be based on s (tjuncture) + f(t) * [s(t) - s(tjuncture)]. where s(tjuncture) is the value ofs (t) corresponding to the final sample of the first block 703. and f (t) is a function that smoothly increases from zero to one. Line 710 shows the value of s(tjuncture) from t = 100 to t = 10 4. Plotted lined 711 shows the time-varying vaiue of s(tjuncture) = f(t) * [s(t) - s (tjuncture)]. where f (t) is a function that smoothly increases from zero to one (e. g.. f (loo) f (101) =.'D. f (102) =. 50. f (103) =. 75. f (104) = 1. 0. etc). In another embodiment of the present invention. f (t) increases from zero to one as follows : 0.. 125, 25.. 50. 1. One of skill in art can appreciate different embodiments of f (t) based on differences in the length of the transition period. the rate of increase from zero to one, the smoothness of the increase from zero to one. differences in logic to implement f (t), etc.

FIG. 8 shows an apparatus in accordance with another embodiment of the present invention. A multiplexer 850 can select between standard output S 851 and transition output T 852 to provide output o (t). Sample stream s (t). in one embodiment, is communicated to standard output S 851. s (t) can be communicated to register 802. which can store the value of the final audio sample immediately prior to the juncture. i. e.. s (tjuncture). In one embodiment. register 80'is a latch register. Subtractor 803 can output the difference between s (t) and s (tjuncture), e.g., s(t) - s(tjuncture). Multiplier 804 can output the product of f (t). which is a : function that can smoothly vary from zero to one over a period of time. and the output oi subtractor 803. Adder 805 can output the sum of the outputs of multiplier 804 and register 802 such that a transition output signal s (liuncture) + f (t) K [S(t) - S(tjuncture)] is communicated to transition output T 852. During typical operations (e.g., when no copied audio sample sequences are spliced into the sample stream to replace a lost audio sample sequence. etc. ;. multiplexer 850 can select the sample stream s (t) present at the standard output S 851 as the output o (t). When a copied audio sample sequence is spliced into the sample stream. the multiplexor 850 can select the transition output signal s (tjuncture) + f (t) * [s (t)-S (tjuctL present at the transition output T 852 during a transition period to smooth a discontinuity between the spliced audio sample sequences. Subsequent to the transition period, the

multiplexor 850 can select select sample stream s (t) present at the standard output S 851 as the output o (t).

For example, referring to FIGS. 7 and 8. multiplexor 850 can output s (t) as output o (t) during the period 0 zu the <-100 b\-outputting the standard outputtin<J the signal After t = 100. multiplexor 850 can output the transition output signal s (tjuncture) + f(t) * [s(t) -s (tjuncturc)] as the output o (t) during the period 100 < t # 104 by outputting output T 852. the signa) present at transition After t = 104, multiplexer 850 can again output s (t) as output o (t) by outputting the signal present at standard output S 851.

FIG. 9 shows an apparatus in accordance with another embodiment of the present invention. An output multiplexer 950 can select between standard output S 951. and transition output T 952 to provide output o (t). In accordance with the embodiment illustrated by FIG. 9. function f (t) can be a function with a dynamic range between zero and one-half. In one embodiment. f (t) smoothly increases from zero to one-half. and then smoothly decreases from one-half to zero.

Sample stream s (t) * in one embodiment, is communicated to standard output S 951. s (t) can be communicated to register 90-'. which can store the value of the final audio sample immediately prior to the juncture, i. e.. s (tjuncture). Subtractor 903 can output the difference between s (t) and s (tjuncture), e. lut.. s (t)-s (tjuncture). Invertor 911 can receive and invert the output of subtractor 903, i. e.. invertor 911 can output s (tjurcturc)-s (t).

First multiplexer 921 can receive each of the outputs of subtractor 903 and inverter 911, and can thereby selectively output one of s (t)-s (tjuncture) and s (tjuncture)-s (t) based on a multiplexer control signal m (t). The value of m (t) can be based at least in part on the value of f (t). m (t) can have a value of logical one (e. g.. five volts. volts. 2. 5 volts, etc.) when f (t) smoothly increases from zero to one-half. and m (t) can have a value of logical zero (e. g.. approximately zero volts, etc.) when f (t) smoothly decreases from one-half to zero. First multiplexer 922 can output s(t) - s(tjuncture) when f (t) smoothly increases from zero to one- half. and output s (tjuncture)-s (t) when f (t) smoothly decreases from one-half to zero.

Multiplier 904 can output the product of f (t) and the output of first multiplexer 9'1.

Second multiplexer 922 can receive selectively output one ofs (tjuncture) and s (t) based on multiplexer control signal m (t). Second multiplexer 92'can output s (tjunciurc) when f (t) smoothly increases from zero to one-half. and output s (t) when f (t) smoothly decreases from one-half to zero.

Adder 905 can output the sum of the outputs of multiplier 904 and second multiplexer 922. When f (t) smoothly increases from zero to one-half. m (t) has a value of logical one and adder 905 can output s (tjuncture) + f (t) * [s (t)-s (tjuncture)]. When f (t) smoothly decreases from one-half to zero, m (t) has a value of logical zero and adder 905 can output s (t) + f (t) * [S (tjuncture)-s (t)].

During typical operations (e. g.. when no copied audio sample sequences are spliced into the sample stream to replace a lost audio sample sequence. etc.). multiplexer 950 can select the sample stream s (t) present at the standard output S 951 as the output o (t). When a copied audio sample sequence is spliced into the sample stream. the multiplexor 950 can select the transition output signal present at transition output T 952 as the output o (t) during both halves of the transition period. The first half of the transition period can coincide witr the period during which f (t) smoothly increases from zero to one-half and m (t) has a value o/ logical one. and the transition output signal present at transition output T 952 is s(tjuncture) - f (t) * [s (i The second half of the transition period can coincide with the period during which f (t) smoothly decreases from one-half to zero. m (t) has a value of logical zero. and the transition output signal present at transition output T 952 is s(t) + f(t) * [s(tjuncture) - s (t)].

In accordance with embodiments of the present described anove. a discontinuity in.. spliced audio stream is smoothed by generating an output signal based on interpolating with. : time-varying weighting function and a reference value based on at least the final sample of the audio sample sequence preceding the juncture having a discontinuity. For example, in one embodiment the interpolation is based at least in part on a delayed audio stream. s (t-x). In another embodiment of the present invention the interpolation is based at least in part on the value of the last sample of the audio sample sequence preceding the juncture having a discontinuitv. It can be appreciated by one of skill in the art that other similar interpolations can be used in accordance with embodiments of the present invention e. (T., an interpolation based at least in part on an average of the last four samples of the audio sample sequence preceding the juncture having a discontinuity, an interpolation based at least in part on a mean of the final samples the last five samples. the number of samples corresponding to the length of the transition period, etc.) of the audio sample sequence preceding the iuncture having a discontinuity, etc.

Embodiments of the present invention include systems incorporating at least one of the apparatus. or performing at least one of the methods. described herein. Examples of systems in accordance with embodiments of the present invention include a digital audio system of a computer. a television. a High Definition television (HDTV) receiver/decoder. compact disc (CD) player, digital versatile disc (DVD) uiaver. Donable audio player, home stereo system, digital radio receiver, etc.

In accordance with embodiments of the present invention described herein, the splice. discontinuous audio streams can be smoothed to avoid audible artifacts (einz pops. clicks. etc.) without modifying the audio sample sequence preceding the juncture having a discontinuity. The points in the audio stream at which splices may occur can be identified in advance at or after a copied audio sample sequence is inserted into the audio stream. etc.). Thus. the value (s) of the last sample (s) of the audio sample sequence preceding the

juncture having a discominuity can be captured, stored, averaged, etc. Methods and apparatus in accordance with embodiments of the present invention advantageously may not require that the need to ramp down an audio sample sequence preceding a discontinuity be identified soon enough to ramp down the preceding sequence. Moreover, ramping the audio stream down to the audio samples of the audio sample seouence subsequent to a discontinuity can reduce the distortion of the spliced stream of audio sample sequences be accordance with one embodiment of the present invention, instructions adapted to be executed by a processor to perform a method in accordance with an embodiment of the present invention are stored on a computer-readable medium and distributed as software. The computer-readable medium can be a device adapted to store digital information. For example. a computer-readable medium includes a portable magnetic disk. such as a floppy disk : or a <BR> <BR> <BR> Zip# disk, manufactured by the lomega Corporation of Roy. Utah: or a Compact Disk Read Only Memory (CD-ROM) as is known in the art for distributing software. The computer- readable medium can be distributed to a user that has a processor suitable for executing instructions adapted to be executed. The term "adapted to be executed" is meant to encompass any instructions that are ready to be executed in their present form (e.g., machine code) by a processor. or require further manipulation (e. g.. compilation. decryption. or provided with an access code. etc.) to be ready to be executed by a processor.

Embodiments of the present invention advantageously allow smoothing of a spliced discontinuous audio stream. The audio sample sequence preceding a discontinuity may not need to be modified (e. g. ramped down to zero, etc.). A reduced amount of distortion of the audio stream can be provided while smoothing the discontinuity. In one embodiment, only the last sample of the audio sample sequence preceding the discontinuity need be captured.

That last sample (e. g.. a point at which a splice occurs) typically can be identified easily in advance.

In the toregoing detailed description, apparatus and methods in accordance with embodiments of the present invention have been described with reference to specific exemplary embodiments. Accordingly. the present specification and figures are to be regarded as illustrative rather than restrictive.