ThomasThebaud
/

PROPS_K16

Feature Extraction

props-composed-gmm-mdn

speaker-embeddings

mixture-density-network

text-to-speaker-embedding

Model card Files Files and versions

ThomasThebaud commited on Sep 14

Commit

d6a4216

·

verified ·

1 Parent(s): a4534ab

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -66,7 +66,7 @@ For reproducible use, review and pin the upstream text-encoder revision in your
 The model was first pretrained to route profile-description embeddings toward components of real per-profile GMMs. It was then fine-tuned with per-utterance Capspeech descriptions by minimizing negative log likelihood of normalized ECAPA speaker embeddings. Model selection used development-set NLL.
-The uploaded artifact is named `16_components_Capspeech_min100_p=1_finetune.pt`, as supplied for this repository. Its saved metadata identifies the parent checkpoint as `...p=1_pretrain.pt` but does not store the fine-tuning fraction. The internal PyTorch archive directory is named `...p=0.1_finetune`; consequently, the precise fine-tuning fraction cannot be established from the checkpoint alone.
 ## Intended uses

 The model was first pretrained to route profile-description embeddings toward components of real per-profile GMMs. It was then fine-tuned with per-utterance Capspeech descriptions by minimizing negative log likelihood of normalized ECAPA speaker embeddings. Model selection used development-set NLL.
+The uploaded artifact is named `16_components_Capspeech_min100_p=1_finetune.pt`, as supplied for this repository.
 ## Intended uses