Yi-Chiao WU

icon indicating a website link https://bigpon.github.io/ icon indicating an email [email protected]

icon company Meta icon location New York City, NY, US icon location Research Scientist @ Meta Reality Labs Research topics: Neural codec Voice Conversion, Speech Synthesis, Speech Enhancement.

Results 23 comments of Yi-Chiao WU
trafficstars

The test results are different from those in the paper

Hi, As I mentioned in point 3, I used the time-warping results of the mcc for f0 alignment. (Directly doing alignment on f0 doesn't make sense. You have to find...

Is it missing some activation functions between some layers?

According to [DAC](https://github.com/descriptinc/descript-audio-codecl), they claimed that snake is much better. However, in AudioDec architecture, we didn't find the tendency. Two possible reasons, 1. Snake is sensitive to initialization and training,...

Table 4 and 5

Hi, 1. It is the processing time in milliseconds (ms), which can be found in the table's caption. 2. No, we don't provide the code to calculate the number of...