The present invention provides an adaptive multi rate encoder and encoding method, the voice activity detection is the synthesis of digital voice encoding frame corresponding, and according to the transmitting type of voice activity detection results and discontinuous transmission of voice compression encoding rate to determine the transmission rate, the change of re made the overall update and the adaptive multi rate encoder and encoding method, encoder according to the type of transmission frame to determine the excitation signal to the decoder synthesized voice signal can accurately reflect the original sound auditory effect. The invention can be directly applied to the voice coding technology of the third generation mobile communication system, the universal mobile communication system.
【技术实现步骤摘要】
【技术保护点】
一种可不连续发送的自适应多速率宽带AMR-WB编码器,在所述AMR-WB编码器中,对输入信号帧进行线性预测,按照话音激活检测结果确定发送类型TX_TYPE,按照所述话音激活检测结果和所述TX_TYPE确定AMR-WB编码帧的编码速率,根据该编码速率为所述输入信号帧编码AMR-WB编码帧,输出类型为TX_TYPE的AMR-WB发送帧,并生成用于编码下一个输入信号帧的所述输入信号帧的激励信号,其特征在于, 由对输入信号帧进行线性预测得到的线性预测参数确定线性预测合成滤波器; 根据一个语音模式编码速率对输入信号帧自适应码书搜索、固定码书搜索所得到的自适应码书参数和固定码书参数生成激励信号,用所述线性预测合成滤波器对该激励信号滤波生成合成数字声音信号帧; 根据对所述合成数字声音信号帧执行的话音激活检测 得到所述的话音激活检测结果; 若所述的话音激活检测结果是有话音,根据按所述的一个语音模式编码速率对输入信号帧进行自适应码书搜索、固定码书搜索所得到的自适应码书参数和固定码书参数为所述输入信号帧编码AMR-WB发送帧,并根据该编码帧中所 使用的基音延迟、 ...
【技术特征摘要】
还没有人留言评论。发表了对其他浏览者有用的留言会获得科技券。