Some questions about inference.py

[JinY97]

Hi, thank you for sharing your work! I have some confusion about it.

• As shown in Eq.(17) in the paper，delta_bar[n] should be equal to (delta_cond[n])* delta[n] / delta[n-1]. However, the code shows that `delta_bar[n] = (delta_cond[n])* delta[n-1]/ delta[n]'.
• The code shows that audio = noisy_audio (line 148), but according to Eq. (10), audio = alpha_cum[-1]**0.5 * noisy_signal + delta[-1] * torch.randn_like(noisy_signal).

Hope to obtain your answer, thank you!

[neillu23]

Hi @JinY97, thank you for your interest in CDiffuSE and your comment!

• The code shows that audio = noisy_audio (line 148), but according to Eq. (10), audio = alpha_cum[-1]**0.5 * noisy_signal + delta[-1] * torch.randn_like(noisy_signal).

This implementation follows the practical technique in A Study on Speech Enhancement Based on Diffusion Probabilistic Model to start with the noisy speech signal. Also, since the alpha_cum[-1] is close to 1 and delta[-1] is close to 0, the starting signal (and the results) is similar to the ones in Eq. (10).

• As shown in Eq.(17) in the paper，delta_bar[n] should be equal to (delta_cond[n])* delta[n] / delta[n-1]. However, the code shows that `delta_bar[n] = (delta_cond[n])* delta[n-1]/ delta[n]'.

I found this would be a typo in the paper, delta_bar[n] should be equal to (delta_cond[n])* delta[n-1]/ delta[n]. I will further update this in the Arxiv paper. Thanks for pointing this out!