Giters

anton-jeran / FAST-RIR

This is the official implementation of our neural-network-based fast diffuse room impulse response generator (FAST-RIR) for generating room impulse responses (RIRs) for a given acoustic environment.

Home Page:https://anton-jeran.github.io/FRIR/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Can not hear the RIR voice

Tim5Tang opened this issue · comments

commented

i have a question, in the web https://anton-jeran.github.io/FRIR/ i can hear the "RIR generated using DAS", but i use your model to generate theRIR-* .wav, i could not hear the voice, how can i check the generated wav is right.

commented

Hi,

My model generates the transfer function (Room Impulse Response) to create reverberant speech. You need to convolve the generated RIR using my model with clean speech to simulate reverberant speech. Once you simulate reverberant speech, you can do a listening test to see whether the generated RIRs are right.

Thanks,

Anton

commented

hi anton,

is there a demo or which script should i use to convolve the generated RIR with clean speech?

thanks!

commented

Hi,

You can find the script to convolve clean speech and monoaural RIR here (https://github.com/anton-jeran/FAST-RIR/blob/main/code_new/augment.py).

You can follow the steps shown here to augment reverberant speech https://github.com/anton-jeran/MULTI-AUDIODEC/tree/main#single-speaker-binaural-speech.

commented

python3 augment.py --speech "clean speech folder" --ir "generated IR folder" --out "output folder" --nthreads 16

commented