REVIEW 12 cited by
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
LibriMix: An Open-Source Dataset for Generalizable Speech Separation
read the original abstract
In recent years, wsj0-2mix has become the reference dataset for single-channel speech separation. Most deep learning-based speech separation models today are benchmarked on it. However, recent studies have shown important performance drops when models trained on wsj0-2mix are evaluated on other, similar datasets. To address this generalization issue, we created LibriMix, an open-source alternative to wsj0-2mix, and to its noisy extension, WHAM!. Based on LibriSpeech, LibriMix consists of two- or three-speaker mixtures combined with ambient noise samples from WHAM!. Using Conv-TasNet, we achieve competitive performance on all LibriMix versions. In order to fairly evaluate across datasets, we introduce a third test set based on VCTK for speech and WHAM! for noise. Our experiments show that the generalization error is smaller for models trained with LibriMix than with WHAM!, in both clean and noisy conditions. Aiming towards evaluation in more realistic, conversation-like scenarios, we also release a sparsely overlapping version of LibriMix's test set.
Forward citations
Cited by 12 Pith papers
-
Ring Mixing with Auxiliary Signal-to-Consistency-Error Ratio Loss for Unsupervised Denoising in Speech Separation
Ring mixing and SCER loss break symmetry in noisy speech separation training, allowing models to learn denoising from noisy mixtures alone and halve residual noise on benchmarks.
-
DASB - Discrete Audio and Speech Benchmark
DASB is a new benchmark for discrete audio tokens showing semantic tokens outperform acoustic ones but discrete representations remain less robust than continuous features across domains.
-
Don't Listen to Me: A Lightweight, Low-Latency Model for Own-Voice Cancellation in Far-Field Speech Enhancement
Introduces own-voice cancellation as a complement to target speaker extraction and benchmarks lightweight 2 ms latency models for far-field speech enhancement.
-
Uncertainty-based Debiasing and Unlearning for Decontamination
UBD leverages ensemble uncertainty to estimate per-sample memorization and construct debiased targets for post-hoc correction or unlearning, yielding output distributions closer to uncontaminated models on MMLU-Pro an...
-
Can Large Audio Language Models Ignore Multilingual Distractors? An Evaluation of Their Selective Auditory Attention Capabilities
Introduces the MUSA benchmark and evaluates LALMs showing that strong single-speaker performance fails to ensure robust selective attention under multilingual interference, with errors from source confusion and unreso...
-
Exploring Token-Space Manipulation in Latent Audio Tokenizers
LATTE creates a compact latent token bottleneck in audio tokenizers that aggregates global information and enables unsupervised editing of attributes like speaker identity via token swapping.
-
Discriminative-Generative Target Speaker Extraction with Decoder-Only Language Models
A hybrid two-stage framework pairs a discriminative front-end for interference suppression with a generative decoder-only LM back-end to improve perceptual quality and speaker consistency in target speaker extraction ...
-
SALMONN: Towards Generic Hearing Abilities for Large Language Models
SALMONN integrates speech and audio encoders with a text-based LLM to process general audio inputs, achieve competitive results on trained tasks, and exhibit emergent cross-modal abilities.
-
Flow Matching-Based Speech Source Separation with Best-of-N Biometric Sampling
A flow-matching speech separator with biometric best-of-N candidate selection and chunk-wise channel alignment achieves competitive separation metrics and the best downstream ASR/SV error rates among evaluated systems...
-
Child-Centric Voice Anonymization in Single and Multi-Speaker Speech via Domain-Adapted SSL Models
Domain adaptation of an SSL voice anonymization pipeline to child speech from the MyST corpus improves intelligibility and quality in single- and multi-speaker settings while preserving privacy.
-
Mind the Gap: Impact of Synthetic Conversational Data on Multi-Talker ASR and Speaker Diarization
Task-dependent simulation strategies for synthetic conversational data allow synthetic-only training to approach real-data baselines for multi-talker ASR and diarization, with mixing yielding further gains.
-
BUT System Description for CHiME-9 MCoRec Challenge
BUT's CHiME-9 MCoRec system conditions Parakeet-v2 ASR on AV-HuBERT visuals for 33.7% WER and uses Qwen3.5 LLM for hierarchical clustering to reach 0.97 F1, beating the baseline by 16.2% WER and 0.15 F1 on the develop...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.