Papers for

conference system engineers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Neural models improve distant speaker diarization in noisy conditions

Neural Multichannel Distant Speaker Diarization with Heavy-tailed Source Separation Model

Abstract: Distant speaker diarization remains challenging due to difficult acoustic environments, varying numbers of speakers and overlapping speech. Model-driven methods are proposed to exploit the speech source features in multi-channel recordings that help diarization. This paper generalizes a neural model that jointly learns to perform blind source separation and diarization over speech mixtures (neural FCASA) with heavy-tailed models. The popular Gaussian distribution has been applied for variance modeling in the original source separation model, which we replace with two families of heavy-tailed models (Leptokurtic Generalized Gaussian distribution and Student's t distribution) to better capture the heavy-tailedness in speech signals. Thanks to the Gaussian scale mixture model, we are able to unify the proposed method and the original one under the same form of learning objective. Our experiments show consistent large improvements in Diarization Error Rate (DER) and Jaccard Error Rate (JER) compared to the baseline on various corpora.

Thu 10 SeptSoundArtificial Intelligence
The gist
Separating and identifying who is speaking in a room with multiple people talking at once and from a distance is hard. The authors improved a method that uses multiple microphones and advanced math models to better separate voices and label who spoke when. They replaced the usual assumptions about voice signal patterns with more flexible ones that better fit real speech. Their experiments show this change reduces mistakes in identifying speakers.
Open 2609.12154v1

Streaming speech recognition balances accuracy speed and memory for multiple speakers

Pushing the Boundaries of Streaming Multi-Speaker ASR: A Systematic Study of Architectural Trade-offs

Abstract: Streaming multi-speaker ASR is a challenging task that must balance accuracy, latency, and efficiency while handling overlapping speech and maintaining coherent long-context modeling over extended conversations in an online fashion. We present a unified framework that categorizes streaming multi-speaker ASR into four architectural strategies based on how diarization and ASR are integrated. Using a shared pair of open-source streaming ASR and diarization models as a common foundation, we derive four multi-speaker ASR systems that differ in whether they employ multiple model instances, fine-tuning, or both. We evaluate these systems across multi-speaker accuracy, single-speaker accuracy degradation, memory footprint, and training complexity. Through this systematic architectural analysis, we clarify the design space for streaming multi-speaker ASR and provide practical guidance for selecting the most suitable approach under diverse deployment constraints.

Wed 9 SeptSound
The gist
Recognizing multiple people talking at the same time in real time is hard because the system has to be fast, accurate, and remember long conversations. The authors studied four different ways to combine identifying who is speaking with transcribing what is said, using the same basic speech and speaker models. They tested how well each way works in accuracy, memory use, and training difficulty. Their study helps clarify which approach is best depending on the situation when building systems that listen to many speakers live.
Open 2609.10265v1