We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
请问README Tabel1 跟Tabel2 中的RNNoise 的MACs 是怎么统计的呢? 单计算网络部分的话,不会有0.04(G/s) 这么大的, 我按你代码中的统计方式,如果按帧长512,帧移256 计算出来的Macs 为5.53M (约0.0055 G/s), 如果按帧长320, 帧移160计算出来的Macs为8.74M (约0.0087 G/s)
The text was updated successfully, but these errors were encountered:
RNNoise的 MACs,使用的是 DeepFilterNet2 的 Table 1 中提供的数据。
Sorry, something went wrong.
https://jmvalin.ca/papers/rnnoise_mmsp2018.pdf 原文中给出的数据,整体的算法复杂度为40Mflops,单网络部分是17.5Mflops , 换成MACs的话是8.75M MACs
你说得对,RNNoise原文把 FFT 和 IFFT 也算进运算量里了。这样我的对比确实存在不严谨的地方,感谢指出。但 RNNoise 只是 GTCRN 的对比对象之一,GTCRN 在 VCTK-DEMAND 数据集上的 PESQ=2.87 是客观成立的。
No branches or pull requests
请问README Tabel1 跟Tabel2 中的RNNoise 的MACs 是怎么统计的呢? 单计算网络部分的话,不会有0.04(G/s) 这么大的, 我按你代码中的统计方式,如果按帧长512,帧移256 计算出来的Macs 为5.53M (约0.0055 G/s), 如果按帧长320, 帧移160计算出来的Macs为8.74M (约0.0087 G/s)
The text was updated successfully, but these errors were encountered: