Convoifilter: A case study of doing cocktail party speech recognition
Paper • 2308.11380 • Published • 1
How to use nguyenvulebinh/voice-filter with Transformers:
# Load model directly
from transformers import VoiceFilter
model = VoiceFilter.from_pretrained("nguyenvulebinh/voice-filter", dtype="auto")The ConVoiFilter model is designed for the purpose of filtering target speaker voices. For a comprehensive understanding of its functionality and methodology, please refer to the associated paper available at: https://arxiv.org/pdf/2308.11380.pdf
For loading and inferring model, you can utilize the provided Colab script. Simply click on the badge below to open the script in Google Colab:
# Load model directly from transformers import VoiceFilter model = VoiceFilter.from_pretrained("nguyenvulebinh/voice-filter", dtype="auto")