| US 6,795,805請求項1 |
US 6,807,524請求項1 |
1. A device for enhancing periodicity of an excitation signal produced in relation to a pitch codevector and an innovative codevector for supplying a signal synthesis filter in view of synthesizing a wideband speech signal, said periodicity enhancing device comprising:
一項用來增強一激發信號週期性之裝置，激發信號之產生與一音調碼向量以及一新創碼向量相關，其為合成一寬頻語音信號提供一信號合成濾波器，前述週期性增強裝置包括 |
1. A perceptual weighting device for producing a perceptually weighted signal in response to a wideband speech signal in order to reduce a difference between the wideband speech signal and a subsequently synthesized wideband speech signal, said perceptual weighting device comprising:
一項感覺加權裝置，用來回應一寬頻語音信號而產生感覺加權信號，藉以減少前述寬頻語音信號與一後續合成之寬頻語音信號間的差異，前述感覺加權裝置包括 |
a) a factor generator (FIG.2 - 204) for calculating a periodicity factor related to the wideband speech signal; and
a)一因子產生器，用來計算出一個與前述寬頻語音信號相關的週期性因子 |
a) a signal preemphasis filter (FIG.1 - 103) responsive to the wideband speech signal for enhancing a high frequency content of the wideband speech signal to thereby produce a preemphasised signal;
a)一信號預加重濾波器，其回應前述寬頻語音信號並用來增強前述寬頻語音信號之高頻分量，藉以產生一預加重信號 |
b) an innovation filter (FIG.2 - 205) for filtering the innovative codevector in relation to said periodicity factor to thereby reduce energy of a low frequency portion of the innovative codevector and enhance periodicity of a low frequency portion of the excitation signal.
b) 一新創濾波器，用來對應前述週期性因子而就新創碼向量進行濾波，藉以減少新創碼向量低頻分量之能量，並增強激發信號低頻分量之週期性 |
b) a synthesis filter calculator (FIG.1 - 104) responsive to said preemphasised signal for producing synthesis filter coefficients; and
b)一合成濾波器計算器，其回應前述預加重信號並用來產生合成濾波器係數 |
 |
c) a perceptual weighting filter (FIG.1 - 105), responsive to said preemphasised signal and said synthesis filter coefficients, for filtering said preemphasised signal in relation to said synthesis filter coefficients to thereby produce said perceptually weighted signal, said perceptual weighting filter having a transfer function with fixed denominator whereby weighting of said wideband speech signal in a formant region is substantially decoupled from a spectral tilt of said wideband speech signal.
c)一感覺加權濾波器，其回應前述預加重信號與前述合成濾波器係數，並用來相對於前述合成濾波器係數、就前述預加重信號進行濾波，藉以產生產生前述感覺加權信號，前述感覺加權濾波器之轉移函數分母為固定，藉以讓在一共振峰區域中所進行之前述寬頻語音信號加權，可以與前述寬頻語音信號之頻譜傾斜分開 |
 |
| US 7,151,802請求項1 |
US 7,260,521請求項1 |
1. A decoder (FIG.2 - 200) for producing a synthesized wideband signal, comprising:
一項用來產生合成寬頻信號之解碼器，包括 |
1. A pitch analysis device for producing a set of pitch codebook parameters, comprising:
一項用來產生音調碼本參數集合之音調分析裝置，包括 |
a) a signal fragmenting device for receiving an encoded version of a wideband signal previously down-sampled during encoding and extracting from said encoded wideband signal version at least pitch codebook parameters, innovative codebook parameters, and linear prediction filter coefficients;
a)一信號分段裝置，用來接收先前在編碼階段被進行縮減取樣之寬頻信號的編碼版本，並從前述寬頻信號編碼版本萃取至少音調碼本參數、新創碼本參數與線性預測濾波器係數 |
a pitch codebook search device (FIG.3 - 301) configured to generate a pitch code vector based on a digitized input audio data, wherein said digitized input audio data represents an input audio signal that has been sampled and digitized;
一音調碼本搜尋裝置，其被配置來基於一數位化輸入音訊資料來產生一音調碼向量，其中前述數位化輸入音訊資料代表一經過取樣與數位化的輸入音訊信號 |
b) a pitch codebook (FIG.2 - 201) responsive to said pitch codebook parameters for producing a pitch codevector;
b)一音調碼本，其回應前述音調碼本參數並用來產生一音調碼向量 |
a) at least two signal paths associated to respective sets of pitch codebook parameters representative of said digitized input audio data,
a)至少兩個與各自相應、代表前述數位化輸入音訊資料之音調碼本參數集合相關的信號路徑， |
c) an innovative codebook (FIG.2 - 218) responsive to said innovative codebook parameters for producing an innovative codevector;
c)一新創碼本，其回應前述新創碼本參數並用來產生一新創碼向量 |
wherein: i) each signal path comprises a pitch prediction error calculating device (FIG.3 - 306) for calculating a pitch prediction error of said pitch codevector from said pitch codebook search device; and其中：
i)個別信號路徑包括、用來計算來自前述音調碼本搜尋裝置之前述音調碼向量之音調預測錯誤的一音調預測錯誤計算裝置；以及 |
d) a combiner circuit for combining said pitch codevector and said innovative codevector to thereby produce an excitation signal;
d)一組合電路，用來組合前述音調碼向量與前述新創碼向量，藉以產生一激發信號 |
ii) at least one of said at least two signal paths comprises a filter (FIG.3 - 305) for filtering the pitch codevector before supplying said pitch codevector to the pitch prediction error calculating device of said at least one signal path; and
ii)前述至少兩個信號路徑中至少一個包括一濾波器，用來在將前述音調碼向量提供給前述至少一個信號路徑之音調預測錯誤計算裝置之前，就音調碼向量進行濾波 |
e) a signal synthesis device including a linear prediction filter for filtering said excitation signal in relation to said linear prediction filter coefficients to thereby produce a synthesized wideband signal, and an oversampler responsive to said synthesized wideband signal for producing an over-sampled signal version of the synthesized wideband signal; and
e)一信號合成裝置，其包括一用來對應前述線性預測濾波器係數、對前述激發信號進行濾波、藉以產生一合成寬頻信號的線性預測濾波器，以及一回應前述合成寬頻信號並用來產生前述合成寬頻信號之超取樣信號版本的超取樣器 |
b) a selector (FIG.3 - 309) for comparing the pitch prediction errors calculated in said at least two signal paths, for choosing the signal path having the lowest calculated pitch prediction error and for selecting the set of pitch codebook parameters associated to the chosen signal path.
b)一選擇器，用來比較在前述至少兩個信號路徑中所計算的音調預測錯誤，藉以選擇具備最低計算音調預測錯誤之信號路徑、以及選擇與被選擇信號路徑相關的音調碼本參數集合 |
f) a high-frequency content recovering device comprising: i) a random noise generator (FIG.2 - 213) for producing a noise sequence having a given spectrum; ii) a spectral shaping unit (FIG.2 - 215) for shaping the spectrum of the noise sequence in relation to linear prediction filter coefficients related to said down-sampled wideband signal; and iii) a signal injection circuit for injecting said spectrally-shaped noise sequence in said over-sampled synthesized signal version to thereby produce said full-spectrum synthesized wideband signal.
f)一高頻分量回復裝置，包括：i)一隨機噪音產生器，用來產生一具備既有頻譜的噪音序列；ii)一頻譜整形元件，用來對應與前述縮減取樣寬頻信號相關之線性預測濾波器係數而就前述噪音序列之頻譜進行整形；以及iii)一信號注入電路，用來將前述經頻譜整形之噪音序列注入前述超取樣合成信號版本中，藉以產生一全頻譜合成寬頻信號 |
 |
 |
| US 7,191,123請求項1 |
1. A method for producing a gain-smoothed codevector during decoding of an encoded wideband signal from a set of wideband signal encoding parameters, said method comprising:
一項用來在解碼經編碼寬頻信號之過程中、從一組寬頻信號編碼參數中產生增益平坦化碼向量之方法，前述方法包括以下步驟 |
finding a codevector in relation to at least one first wideband signal encoding parameter of said set;
找出與前述組中至少一個第一寬頻信號編碼參數相關的碼向量 |
calculating a first factor representative of voicing in the wideband signal in response to at least one second wideband signal encoding parameter of said set;
計算出代表寬頻信號中發聲的第一因子，以回應前述組中至少一個第二寬頻信號編碼參數 |
calculating a second factor representative of stability of said wideband signal in response to at least one third wideband signal encoding parameter of said set;
計算出代表前述寬頻信號中穩定性的第二因子，以回應前述組中至少一個第三寬頻信號編碼參數 |
calculating a smoothing gain based on said first and second factors; and
依據前述第一與第二因子計算出一平坦化增益 |
amplifying the found codevector with said smoothing gain to thereby produce said gain-smoothed codevector.
以前述平坦化增益放大所找出的碼向量，藉以產生前述增益平坦化碼向量 |
 |