import onnxruntime as ort from transformers import HubertModel import torch
Source du modèle
Description de la source
import onnxruntime as ort
from transformers import HubertModel
import torch
# Load the upstream HuBERT model.
upstream = HubertModel.from_pretrained("facebook/hubert-base-ls960")
upstream.eval()
# Load the autoencoder model.
postprocessor = ort.InferenceSession("sslzip_16.onnx")
node_name = postprocessor.get_inputs()[0].name
# Prepare an input waveform (assuming 16kHz audio).
x = torch.randn(1, 16000)
# Extract the latent representation for downstream tasks.
with torch.inference_mode():
h = upstream(x, output_hidden_states=True).hidden_states[-1]
z = postprocessor.run(None, {node_name: h.cpu().numpy()})[0]
# Use z as you like.
print(z.shape)
The pretrained model was developed using the LibriSpeech corpus and is distributed under the same license (CC BY 4.0). Please include credit to Nagoya Institue of Technology and Techno-Speech, Inc. when using this model.
Sources
1 sourceVérifié 5 sept.
Artefacts du modèle
1 artefactExtraits de sources
2 extraits@InProceedings{yoshimura2025sslzip,
author = {Takenori Yoshimura and Shinji Takaki and Kazuhiro Nakamura and Keiichiro Oura and Takato Fujimoto and Kei Hashimoto and Yoshihiko Nankaku and Keiichi Tokuda},
title = {{SSLZip}: Simple autoencoding for enhancing self-supervised speech representations in speech generation},
booktitle = {13th ISCA Speech Synthesis Workshop (SSW 2025)},
pages = {xxx--xxx},
year = {2025},
}
--- license: cc-by-4.0 datasets: - openslr/librispeech_asr language: - en pipeline_tag: audio-to-audio --- # SSLZip ## Usage ```py import onnxruntime as ort from transformers import HubertModel import torch # Load the upstream HuBERT model. upstream = HubertModel.from_pretrained("facebook/hubert-base-ls960") upstream.eval() # Load the autoencoder model. postprocessor = ort.InferenceSession("sslzip_16.onnx") node_name = postprocessor.get_inputs()[0].name # Prepare an input waveform (assuming 16kHz audio). x = torch.randn(1, 16000) # Extract the latent representation for downstream tasks. with torch.inference_mode(): h = upstream(x, output_hidden_states=True).hidden_states[-1] z = postprocessor.run(None, {node_name: h.cpu().numpy()})[0] # Use z as you like. print(z.shape) ``` ## License The pretrained model was developed using the LibriSpeech corpus and is distributed under the same license (CC BY 4.0). Please include credit to Nagoya Institue of Technology and Techno-Speech, Inc. when using this model. ## Citation ```bibtex @InProceedings{yoshimura2025sslzip, author = {Takenori Yoshimura and Shinji Takaki and Kazuhiro Nakamura and Keiichiro Oura and Takato Fujimoto and Kei Hashimoto and Yoshihiko Nankaku and Keiichi Tokuda}, title = {{SSLZip}: Simple autoencoding for enhancing self-supervised speech representations in speech generation}, booktitle = {13th ISCA Speech Synthesis Workshop (SSW 2025)}, pages = {xxx--xxx}, year = {2025}, } ```
Source context: 0 downloads · 0 likes · Pipeline audio-to-audio · Repo takenori-y/SSLZip-16