摘要:
An embodiment of a synthesis filterbank (800) for filtering a plurality of input frames, wherein each input frame comprises M ordered input values y k (0), ..., y k (M-1), comprises an inverse type-IV discrete cosine transform frequency/time converter (810) for providing a plurality of output frames, an output frame comprising 2M ordered output samples x k (0), ..., x k (2M-1), a windower (820) for generating a plurality of windowed frames, a windowed frame comprising a plurality of windowed samples z k (0), ..., z k (2M-1), an overlap/adder (820) for providing an intermediate frame comprising a plurality of intermediate samples m k (0), ..., m k (M-1) and a lifter (830) for providing an added frame comprising a plurality of added samples out k (0), ..., out k (M-1) based on the equation out k n = m k n + l n - M / 2 ⋅ m k - 1 M - 1 - n for n = M / 2 , … , M - 1 and out k n = m k n + l M - 1 - n ⋅ out k - 1 M - 1 - n for n = 0 , … , M / 2 - 1 , wherein 1(0), ...,1(M-1) are real-valued lifting coefficients.
摘要:
A speech decoding method according to a code-excited linear prediction (CELP) speech. Obtaining a time series vector from an excitation codebook and determining according to a decoded gain if a modification of said vector is necessary. If required the vector is modified such that its number of zero - amplitude valued samples is changed.
摘要:
In a CELP type speech coder, the excitation is quantized by vectors from a random codebook. The random codebook is made of a fixed waveform storage section (181), followed by a vector rearranging unit (182). The rearranging section (182) shifts the vectors to positions determined to minimize the quantization distorsion using a pulse placement methodology of an algebraic coder. The vectors are summed (183) to generate the excitation code vector.
摘要:
In a CELP type speech coder, the excitation is quantized by vectors from a random codebook. The random codebook is made of a fixed waveform storage section (181), followed by a vector rearranging unit (182). The rearranging section (182) shifts the vectors to positions determined to minimize the quantization distorsion using a pulse placement methodology of an algebraic coder. The vectors are summed (183) to generate the excitation code vector.
摘要:
A random code vector reading section and a random codebook of a conventional CELP type speech coder/decoder are respectively replaced with an oscillator for outputting different vector streams in accordance with values of input seeds, and a seed storage section for storing a plurality of seeds . This makes it unnecessary to store fixed vectors as they are in a fixed codebook (ROM), thereby considerably reducing the memory capacity.
摘要:
A CELP speech coder or decoder comprising an adaptive codebook (223) for generating an adaptive code vector; a random codebook (221,222), for generating a random code vector; a synthesis filter (215) for receiving a random code vector generated from the random codebook (221,222) so as to perform LPC synthesis; wherein said random codebook (221,222) is formed by an excitation vector generator comprising fixed waveform arranging means (182) for arranging one or more fixed waveforms (v1,v2,v3) stored in a fixed waveform storage means (181) in accordance with the positions (P1,P2,P3) and polarities of an input impulse vector when the input speech is determined to be unvoiced.
摘要:
In a CELP type speech coder, the excitation is quantized by vectors from a random codebook. The random codebook is made of a fixed waveform storage section (181), followed by a vector rearranging unit (182). The rearranging section (182) shifts the vectors to positions determined to minimize the quantization distorsion using a pulse placement methodology of an algebraic coder. The vectors are summed (183) to generate the excitation code vector.
摘要:
In a CELP type speech coder, the excitation is quantized by vectors from a random codebook. The random codebook is made of a fixed waveform storage section (181), followed by a vector rearranging unit (182). The rearranging section (182) shifts the vectors to positions determined to minimize the quantization distorsion using a pulse placement methodology of an algebraic coder. The vectors are summed (183) to generate the excitation code vector.