Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2611.09368

AISHELL8-FISHEYE: A Fisheye Audio-Visual Dataset for Target Speaker Extraction with Distortion-Aware Baselines

Peijun Yang·Zhan Jin·Juan Liu·Hui Bu·Xiaoyi Qin·Ruyi Gan·Hao Wang·Ming Li
View original source ↗Read source document ↗Export metadata ↓

Abstract

Fisheye cameras offer wide visual coverage but distort facial and lip geometry, challenging audio-visual target speaker extraction. We introduce AISHELL8-FISHEYE, a 47.03-hour corpus of 94 Mandarin speakers with synchronized audio and paired fisheye and normal-camera videos. Speaker-disjoint splits and synthetic two-speaker mixtures support reproducible evaluation under matched acoustic conditions. We also establish a baseline that combines a lip-reading encoder with a frozen distortion-aware branch transferred from fisheye head pose estimation. On fisheye video, adding this branch to an adapted lip encoder improves scale-invariant signal-to-distortion ratio improvement from 10.47 to 11.84 dB. The branch does not improve performance with a frozen lip encoder, highlighting the importance of visual adaptation. A substantial gap to paired normal-camera inputs remains. The corpus and baseline resources support further research on speech extraction from distorted visual cues.

Explore related subjects

eess.AS

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Peijun Yang, Zhan Jin, Juan Liu, Hui Bu, Xiaoyi Qin, Ruyi Gan, Hao Wang, Ming Li. 2026-11-06. AISHELL8-FISHEYE: A Fisheye Audio-Visual Dataset for Target Speaker Extraction with Distortion-Aware Baselines. https://arxiv.org/abs/2611.09368

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

REAL-2MIX: Evaluating Identity Faithfulness in Audio-Visual Target Speaker Extraction

Audio-visual target speaker extraction should recover the person indicated by the video, rather than repeatedly return the acoustically dominant voice. We introduce REAL-2MIX, a Mandarin benchmark of jointly recorded two-speaker mixtures with synchronized multi-view video and scene-matched solo references. Its 77 scenes contain 7,598 clips; filtering yields 6,038 evaluable mixtures, each queried for both speakers, giving 12,076 paired target queries. We also curate VOXBLINK2-AVSE, comprising 250,828 synchronized audio--lip pairs. A seven-system TF-GridNet study examines visual representations, training corpora, and conditioning strategies. We evaluate transcription accuracy, speaker identity, residual overlap, and perceptual quality. The AVSR early-fusion system achieves 22.99% CER, and 93.38% of its outputs are more similar to the target speaker than to the interferer. Yet only 42.49% satisfy both this identity criterion and the overlap detector's single-speaker criterion; both target queries satisfy these criteria for just 20.77% of mixtures. A separate Real-AVSE challenge study examines development-remix adaptation. The benchmark reveals that favorable content and identity scores need not coincide with low detected overlap, motivating paired-target evaluation.

eess.AS↗
arXiv2026

AccentCanvas: Activity-Guided Bidirectional Accent Conversion

Accent conversion for dubbing must change pronunciation while preserving content, voice, and source timing. Normalization alone cannot produce requested non-native accents, and matching total duration does not ensure local speech/pause alignment. We present AccentCanvas, a shared discrete diffusion model for normalization and L1-to-L2 conversion, trained on cross-paired shared-prompt recordings and synthetic pairs. Frame-level speech activity and repetition-preserving units guide local timing through synthesis. Across both tasks, AccentCanvas achieves the highest accent conversion performance (Accent Arrival) and mean listener-rated accent similarity among compared systems, with over 73% lower speech-boundary error than the best baseline for each task. These gains accompany higher L1-to-L2 WER than TokAN and Vevo-Style, highlighting a tradeoff between requested-accent resemblance and ASR-based content recognition. https://smokedindia.github.io/accentcanvas/

eess.AS↗
arXiv2026

GTCRN-SS: Lightweight Causal Speech Separation with Residual Allocation and Reconstruction Refinement

Low-complexity causal speech separation must preserve speaker information and allocate residual mixture energy without relying on future feature frames. We present GTCRN-SS, a two-stage time-frequency separator that adapts the compact GTCRN enhancement backbone to fixed-output speech separation. The model combines internal residual allocation, reconstruction-conditioned refinement, and frame-aligned recurrent-state transfer. Its complex-mask estimator assigns non-target content to an unreturned residual branch, while the complementary speech estimate conditions a second stage whose final masks remain anchored to the original mixture. A lightweight low-rank adapter transfers causal recurrent evidence between recursive executions without adding another recurrent memory network. The neural graph excludes future feature frames and incurs only the fixed analysis look-ahead of the centered short-time Fourier transform. On noisy Libri2Mix, GTCRN-SS achieves 10.1 dB SDRi and 9.5 dB SI-SNRi with fewer than one million parameters and a low computational footprint. Independently trained ablations confirm the benefits of recurrent-state transfer and reconstruction conditioning, and corpus-level diagnostics show that the internal branch primarily captures residual rather than target-speech content. These results establish GTCRN-SS as an effective quality-efficiency operating point for causal monaural speech separation.

eess.AS↗