Hey guys,
I bought a pair of Zastone zt-9908 radios. For those of you not familiar with them, they are digital radios using the dpmr standard. I did a little digging (ok, a lot of digging) and found some dirt on the vocoder used. It uses a new flavor of vocoder developed by a few people from Tsinghua University. The vocoder is called ASELP. It is more specifically, Advanced Sinusoidal Excitation Linear Prediction.
This vocoder apparently outperforms the ambe+2 vocoder.
And for those of you interested, I have found the university research document with the vocoders' guts. I'm not sure if it is what can be used to create a decoder, but the algorithms are there.
And as far as I know, the vocoder ambe+2c (not the actual ambe+2) is actually just ASELP.
But hopefully, we can get some use out of this for something, so I'm attaching the pdf of the vocoder. *warning it is in Chinese.
I bought a pair of Zastone zt-9908 radios. For those of you not familiar with them, they are digital radios using the dpmr standard. I did a little digging (ok, a lot of digging) and found some dirt on the vocoder used. It uses a new flavor of vocoder developed by a few people from Tsinghua University. The vocoder is called ASELP. It is more specifically, Advanced Sinusoidal Excitation Linear Prediction.
This vocoder apparently outperforms the ambe+2 vocoder.
And for those of you interested, I have found the university research document with the vocoders' guts. I'm not sure if it is what can be used to create a decoder, but the algorithms are there.
And as far as I know, the vocoder ambe+2c (not the actual ambe+2) is actually just ASELP.
But hopefully, we can get some use out of this for something, so I'm attaching the pdf of the vocoder. *warning it is in Chinese.