trends.barnas.net

model · first seen 4 Oct 2026

speech foundation model

Large self-supervised speech encoders such as Wav2Vec2, HuBERT, and WavLM adapted to downstream tasks.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

adversarial robustnessspeech tokenization