-
Hello,
I have a question about SE models. I want to get DCCRN single channel model that has the best pesq value for CHiMe4 dataset but couldn't see it in Hugging Face models page.
Thanks.
-
## About this issue
I open this issue to start a discussion about some limitations that I have encountered when using Scaper for speech recognition in noisy environments.
## General limitations
…
-
### Describe the bug
Speech enhancement not working on local audio file
### Expected behaviour
it should run and generate result, as transcribe is working fine for same path
### To Reproduce
impo…
-
### Discussed in https://github.com/langchain-ai/langchain/discussions/27404
Originally posted by **kodychik** October 16, 2024
### Checked
- [X] I searched existing ideas and did not find …
-
Dear Dr. Nursadul Mamun，
Thank you for your work, it looks stunning! I replaced your dataset with VoiceBank + DEMAND and had great results as well. Would you be interested in posting the results …
-
Hi!
Thank you for your very interesting research regarding the application of Mamba to Speech Enhancement! Mamba models are very exciting, and their application to Sound-related tasks seem to be a …
-
Can this architecture recover speech from the noisy speech?
-
## Goal
- Remove the need to press the button, detect the voice
- Medium-term
- Enables ambient voice detection
- Enables interruptibility
- Small model that has binary classifier for vo…
-
### Your idea
Сurrently the most actual version of Audacity 3.5 and it uses libopus v1.4 - a year old codec library, is there any chance to update to a more latest version of libopus v1.5.2 from Apr …
-
#### Is this a bug, enhancement, or feature request?
Bug
#### Describe your proposal.
**Observed Behavior:**
As soon as the user is redirected to the dialog the speech output is made as 'Dia…