Downloads · 30 days
0
d_vector_yellow_inherent_elephant
boeddeker/d_vector_yellow_inherent_elephant
d_vector_yellow_inherent_elephant is a machine learning model from boeddeker. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
This repository contains a ResNet34-based d-vector system from [1] to extract speaker embeddings trained on VoxCeleb augmented with MUSAN and simulated room impulse responses.
Access
Public
Updated Jun 4, 2024
Repo size
49.4 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth49.4 MB · 100%
At a glance
- License
- mit
- Access
- Public
- Created
- Jun 4, 2024
- Updated
- Jun 4, 2024
- SHA
- bc73ddd7
Model card
From the Hugging Face model README
This repository contains a ResNet34-based d-vector system from [1] to extract speaker embeddings trained on VoxCeleb augmented with MUSAN and simulated room impulse responses.
It was used for [2] and [3]. It is used in [4], which contains the code for [3]. See [4], to get an example, how you can use this speaker embedding extractor.
- [1] T. Cord-Landwehr, C. Boeddeker, C. Zorila, R. Doddipatla, and ˘ R. Haeb-Umbach, “Frame-wise and overlap-robust speaker embeddings for meeting diarization,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023, pp. 1–5.
- [2] C. Boeddeker, T. Cord-Landwehr, T. von Neumann, and R. Haeb- Umbach, “Multi-stage diarization refinement for the CHiME-7 DASR scenario,” in Proc. CHiME, 2023, pp. 51–56.
- [3] C. Boeddeker, T. Cord-Landwehr, and R. Haeb- Umbach, “Once more Diarization: Improving meeting transcription systems through segment-level speaker reassignment,” accepted for ISCA Interspeech,
- License
- mit
- Created
- Jun 4, 2024
- Updated
- Jun 4, 2024