# TripleC Learning and Lightweight Speech Enhancement for Multi-Condition Target Speech Extraction

arXiv: https://arxiv.org/abs/2512.04945 · HTML: https://mlsanity.com/paper/2512.04945/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Speech enhancement and noise suppression, one of 11 directions in Speech and audio language models.
- Builds on 22 of its 32 references that are in the map.
- First posted on arXiv 4 Dec 2025.

## Placed in

- Speech and audio language models / [Speech enhancement and noise suppression](https://mlsanity.com/speech/speech-enhancement-and-noise-suppression/)

## Builds on (most cited first, 22 of 22)

- [SDR - half-baked or well done?](https://mlsanity.com/paper/1811.02508/) (2018)
- [LibriMix: An Open-Source Dataset for Generalizable Speech Separation](https://mlsanity.com/paper/2005.11262/) (2020)
- [ICASSP 2022 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2202.13288/) (2022)
- [ICASSP 2023 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2303.11510/) (2023)
- [ICASSP 2021 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2009.06122/) (2020)
- [AnyEnhance: A Unified Generative Model with Prompt-Guidance and Self-Critic for Voice Enhancement](https://mlsanity.com/paper/2501.15417/) (2025)
- [High Fidelity Speech Enhancement with Band-split RNN](https://mlsanity.com/paper/2212.00406/) (2022)
- [Metis: A Foundation Speech Generation Model with Masked Generative Pre-training](https://mlsanity.com/paper/2502.03128/) (2025)
- [X-SepFormer: End-to-end Speaker Extraction Network with Explicit Optimization on Speaker Confusion](https://mlsanity.com/paper/2303.05023/) (2023)
- [DPCCN: Densely-Connected Pyramid Complex Convolutional Network for Robust Speech Separation And Extraction](https://mlsanity.com/paper/2112.13520/) (2021)
- [DDTSE: Discriminative Diffusion Model for Target Speech Extraction](https://mlsanity.com/paper/2309.13874/) (2023)
- [Target Speaker Extraction by Directly Exploiting Contextual Information in the Time-Frequency Domain](https://mlsanity.com/paper/2402.17146/) (2024)
- [USEF-TSE: Universal Speaker Embedding Free Target Speaker Extraction](https://mlsanity.com/paper/2409.02615/) (2024)
- [Personalized Speech Enhancement Without a Separate Speaker Embedding Model](https://mlsanity.com/paper/2406.09928/) (2024)
- [Multi-Level Speaker Representation for Target Speaker Extraction](https://mlsanity.com/paper/2410.16059/) (2024)
- [The NPU-Elevoc Personalized Speech Enhancement System for ICASSP2023 DNS Challenge](https://mlsanity.com/paper/2303.06811/) (2023)
- [SEF-PNet: Speaker Encoder-Free Personalized Speech Enhancement with Local and Global Contexts Aggregation](https://mlsanity.com/paper/2501.11274/) (2025)
- [A Framework for Unified Real-time Personalized and Non-Personalized Speech Enhancement](https://mlsanity.com/paper/2302.11768/) (2023)
- [An Investigation on Speaker Augmentation for End-to-End Speaker Extraction](https://mlsanity.com/paper/2505.21805/) (2025)
- [Dynamic Acoustic Compensation and Adaptive Focal Training for Personalized Speech Enhancement](https://mlsanity.com/paper/2211.12097/) (2022)
- [Unified Architecture and Unsupervised Speech Disentanglement for Speaker Embedding-Free Enrollment in Personalized Speech Enhancement](https://mlsanity.com/paper/2505.12288/) (2025)
- [Lightweight speech enhancement guided target speech extraction in noisy multi-speaker scenarios](https://mlsanity.com/paper/2508.19583/) (2025)

## Abstract (arXiv)

In our recent work, we proposed Lightweight Speech Enhancement Guided Target Speech Extraction (LGTSE) and demonstrated its effectiveness in multi-speaker-plus-noise scenarios. However, real-world applications often involve more diverse and complex conditions, such as one-speaker-plus-noise or two-speaker-without-noise. To address this challenge, we extend LGTSE with a Cross-Condition Consistency learning strategy, termed TripleC Learning. This strategy is first validated under multi-speaker-plus-noise condition and then evaluated for its generalization across diverse scenarios. Moreover, building upon the lightweight front-end denoiser in LGTSE, which can flexibly process both noisy and clean mixtures and shows strong generalization to unseen conditions, we integrate TripleC learning with a proposed parallel universal training scheme that organizes batches containing multiple scenarios for the same target speaker. By enforcing consistent extraction across different conditions, easier cases can assist harder ones, thereby fully exploiting diverse training data and fostering a robust universal model. Experimental results on the Libri2Mix three-condition tasks demonstrate that the proposed LGTSE with TripleC learning achieves superior performance over condition-specific models, highlighting its strong potential for universal deployment in real-world speech applications.