@inproceedings{sharma-gupta-2026-early,
title = "Early Warning for Target-Speaker Extraction: Reliability under Distribution and Failure-Definition Shift",
author = "Sharma, Shubham and
Gupta, Manik",
editor = "Shekhar, Ravi and
Swain, Monorama and
Mishra, Jagabandhu and
Dhar, Sandipan and
Mouratidis, Haralambos and
Purver, Matthew",
booktitle = "Proceedings of the {IMPACT}-{SPEECH} Workshop: Identifying, Measuring, Preventing, and Assessing Consequences of Bias in Speech {LLM}s",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.impactspeech-1.4/",
pages = "33--41",
ISBN = "979-8-89176-479-8",
abstract = "Target-speaker extraction (TSE) systems are typically evaluated only after the complete output has been produced, limiting the ability to identify potentially unreliable outputs during generation. We investigate whether failure risk can instead be estimated causally from only the first three seconds of the emitted TSE waveform. We develop an early-warning monitor using 11,946 outputs spanning five conversational corpora and six TSE systems. Because independent failure labels are unavailable for this discovery cohort, we define a proxy adverse outcome from seven complete-output indicators of degraded or atypical TSE behaviour. After excluding monitoring features that overlap with the proxy construction, the monitor predicts this eventual outcome from the early waveform with an AUROC of 0.8062 and an AUPRC of 0.6606. At a 15{\%} review budget, it identifies 54.47{\%} of proxy-adverse outputs with 66.29{\%} precision. To test whether this early-warning signal transfers beyond the discovery setting, we freeze the monitor and evaluate it unchanged on 2,000 unseen outputs with independently computed SI-SDRi. When failure is defined as SI-SDRi {\ensuremath{\leq}} 0 dB, performance falls to near chance (AUROC 0.5354; AUPRC 0.0700). Yet, on the same external outputs, the frozen predictions remain strongly associated with the original proxy outcome (AUROC 0.8092; AUPRC 0.8481). The proxy and SI-SDRi failure definitions agree only weakly ( {\ensuremath{\kappa}} = 0 .028, MCC= 0 .064), while substantial featuredistribution shift is observed between the discovery and external cohorts (45 of 63 features have PSI {\ensuremath{\geq}} 0.25). These results show that short-prefix monitoring can provide useful early risk prioritization within the setting in which the monitor is developed, but that such signals do not automatically transfer to independently defined failure outcomes. Reliable deployment therefore requires explicit attention to both distribution shift and the definition of failure."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sharma-gupta-2026-early">
<titleInfo>
<title>Early Warning for Target-Speaker Extraction: Reliability under Distribution and Failure-Definition Shift</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shubham</namePart>
<namePart type="family">Sharma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manik</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the IMPACT-SPEECH Workshop: Identifying, Measuring, Preventing, and Assessing Consequences of Bias in Speech LLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ravi</namePart>
<namePart type="family">Shekhar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Monorama</namePart>
<namePart type="family">Swain</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jagabandhu</namePart>
<namePart type="family">Mishra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandipan</namePart>
<namePart type="family">Dhar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haralambos</namePart>
<namePart type="family">Mouratidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Purver</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-479-8</identifier>
</relatedItem>
<abstract>Target-speaker extraction (TSE) systems are typically evaluated only after the complete output has been produced, limiting the ability to identify potentially unreliable outputs during generation. We investigate whether failure risk can instead be estimated causally from only the first three seconds of the emitted TSE waveform. We develop an early-warning monitor using 11,946 outputs spanning five conversational corpora and six TSE systems. Because independent failure labels are unavailable for this discovery cohort, we define a proxy adverse outcome from seven complete-output indicators of degraded or atypical TSE behaviour. After excluding monitoring features that overlap with the proxy construction, the monitor predicts this eventual outcome from the early waveform with an AUROC of 0.8062 and an AUPRC of 0.6606. At a 15% review budget, it identifies 54.47% of proxy-adverse outputs with 66.29% precision. To test whether this early-warning signal transfers beyond the discovery setting, we freeze the monitor and evaluate it unchanged on 2,000 unseen outputs with independently computed SI-SDRi. When failure is defined as SI-SDRi \ensuremathłeq 0 dB, performance falls to near chance (AUROC 0.5354; AUPRC 0.0700). Yet, on the same external outputs, the frozen predictions remain strongly associated with the original proxy outcome (AUROC 0.8092; AUPRC 0.8481). The proxy and SI-SDRi failure definitions agree only weakly ( \ensuremathąppa = 0 .028, MCC= 0 .064), while substantial featuredistribution shift is observed between the discovery and external cohorts (45 of 63 features have PSI \ensuremath\geq 0.25). These results show that short-prefix monitoring can provide useful early risk prioritization within the setting in which the monitor is developed, but that such signals do not automatically transfer to independently defined failure outcomes. Reliable deployment therefore requires explicit attention to both distribution shift and the definition of failure.</abstract>
<identifier type="citekey">sharma-gupta-2026-early</identifier>
<location>
<url>https://aclanthology.org/2026.impactspeech-1.4/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>33</start>
<end>41</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Early Warning for Target-Speaker Extraction: Reliability under Distribution and Failure-Definition Shift
%A Sharma, Shubham
%A Gupta, Manik
%Y Shekhar, Ravi
%Y Swain, Monorama
%Y Mishra, Jagabandhu
%Y Dhar, Sandipan
%Y Mouratidis, Haralambos
%Y Purver, Matthew
%S Proceedings of the IMPACT-SPEECH Workshop: Identifying, Measuring, Preventing, and Assessing Consequences of Bias in Speech LLMs
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-479-8
%F sharma-gupta-2026-early
%X Target-speaker extraction (TSE) systems are typically evaluated only after the complete output has been produced, limiting the ability to identify potentially unreliable outputs during generation. We investigate whether failure risk can instead be estimated causally from only the first three seconds of the emitted TSE waveform. We develop an early-warning monitor using 11,946 outputs spanning five conversational corpora and six TSE systems. Because independent failure labels are unavailable for this discovery cohort, we define a proxy adverse outcome from seven complete-output indicators of degraded or atypical TSE behaviour. After excluding monitoring features that overlap with the proxy construction, the monitor predicts this eventual outcome from the early waveform with an AUROC of 0.8062 and an AUPRC of 0.6606. At a 15% review budget, it identifies 54.47% of proxy-adverse outputs with 66.29% precision. To test whether this early-warning signal transfers beyond the discovery setting, we freeze the monitor and evaluate it unchanged on 2,000 unseen outputs with independently computed SI-SDRi. When failure is defined as SI-SDRi \ensuremathłeq 0 dB, performance falls to near chance (AUROC 0.5354; AUPRC 0.0700). Yet, on the same external outputs, the frozen predictions remain strongly associated with the original proxy outcome (AUROC 0.8092; AUPRC 0.8481). The proxy and SI-SDRi failure definitions agree only weakly ( \ensuremathąppa = 0 .028, MCC= 0 .064), while substantial featuredistribution shift is observed between the discovery and external cohorts (45 of 63 features have PSI \ensuremath\geq 0.25). These results show that short-prefix monitoring can provide useful early risk prioritization within the setting in which the monitor is developed, but that such signals do not automatically transfer to independently defined failure outcomes. Reliable deployment therefore requires explicit attention to both distribution shift and the definition of failure.
%U https://aclanthology.org/2026.impactspeech-1.4/
%P 33-41
Markdown (Informal)
[Early Warning for Target-Speaker Extraction: Reliability under Distribution and Failure-Definition Shift](https://aclanthology.org/2026.impactspeech-1.4/) (Sharma & Gupta, IMPACT-SPEECH 2026)
ACL