https://yuiga.dev/blog/en/tags/%E9%9F%B3%E5%A3%B0/ Recent content in 音声 on 行李の底に収めたり[YuWd] Hugo -- gohugo.io en ©2026, All Rights Reserved Thu, 12 May 2022 17:41:56 +0900 https://yuiga.dev/blog/en/ja/posts/adapt-and-adjust_overcoming_the_long-tail_problem_of_multilingual_speech_recognition/ Thu, 12 May 2022 17:41:56 +0900 Thu, 12 May 2022 17:41:56 +0900 https://yuiga.dev/blog/en/ja/posts/adapt-and-adjust_overcoming_the_long-tail_problem_of_multilingual_speech_recognition/ Adapt-and-Adjust (A2), end-to-endの多言語音声認識モデル multilingual language modelをspeach-decoderとする Dual-Adaptersを採用 言語ごとに特徴抽出器を切り替えるイメージ これってほんとに言語ごとに切り替わってるの？ Adapterは Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Modelが初出？ → と思ったら違った 初出: Learning multiple visual domains with residual adapters YuWd (Yuiga Wada) featured image 論文 不均衡データ 音声