中文标题#
語音分離中帶有噪聲參考的尺度不變信干比研究
英文标题#
A Study of the Scale Invariant Signal to Distortion Ratio in Speech Separation with Noisy References
中文摘要#
本文研究了在監督語音分離中，當訓練參考包含噪聲時（如事實上的基準數據集 WSJ0-2Mix），將尺度不變信噪比（SI-SDR）用作評估和訓練目標的含義。對帶有噪聲參考的 SI-SDR 的推導表明，噪聲會限制可達到的 SI-SDR，或導致分離輸出中出現不需要的噪聲。為了解決這個問題，提出了一種增強參考的方法，並使用 WHAM! 對混合信號進行增強，旨在訓練避免學習噪聲參考的模型。在這些增強數據集上訓練的兩個模型使用非侵入式 NISQA.v2 度量進行評估。結果表明分離語音中的噪聲減少，但表明處理參考可能會引入伪影，限制整體質量的提升。在 WSJ0-2Mix 和 Libri2Mix 測試集上，所有模型的 SI-SDR 與感知噪聲之間存在負相關，這支持了從推導中得出的結論。
英文摘要#
This paper examines the implications of using the Scale-Invariant Signal-to-Distortion Ratio (SI-SDR) as both evaluation and training objective in supervised speech separation, when the training references contain noise, as is the case with the de facto benchmark WSJ0-2Mix. A derivation of the SI-SDR with noisy references reveals that noise limits the achievable SI-SDR, or leads to undesired noise in the separated outputs. To address this, a method is proposed to enhance references and augment the mixtures with WHAM!, aiming to train models that avoid learning noisy references. Two models trained on these enhanced datasets are evaluated with the non-intrusive NISQA.v2 metric. Results show reduced noise in separated speech but suggest that processing references may introduce artefacts, limiting overall quality gains. Negative correlation is found between SI-SDR and perceived noisiness across models on the WSJ0-2Mix and Libri2Mix test sets, underlining the conclusion from the derivation.
文章页面#
PDF 获取#
抖音掃碼查看更多精彩內容