@inproceedings{sheikh-ali-etal-2026-menaspeechbank,
title = "{MENAS}peech{B}ank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for {S}peech{LLM}s",
author = "Sheikh Ali, Zien and
Hassan Bhatti, Hunzalah and
Nandi, Rabindra Nath and
Chowdhury, Shammur Absar and
Alam, Firoj",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2700/",
pages = "61422--61444",
ISBN = "979-8-89176-459-0",
abstract = "Audio large language models (AudioLLMs) enable instruction following over speech and general audio, but progress is limited by the scarcity of *diverse, conversational, and instruction-aligned* speech-text data. This gap is particularly pronounced for persona-grounded and dialectal interactions, where collecting real multi-speaker recordings remains costly and slow. We introduce **MENASpeechBank**, a reference speech bank comprising **$\sim$18K high-quality utterances** from **124 speakers** spanning multiple MENA countries, covering English, Modern Standard Arabic (MSA), and regional Arabic varieties. We develop a controllable data pipeline that *(i)* constructs persona profiles enriched with World Values Survey (WVS) inspired attributes, *(ii)* defines a taxonomy driven $\sim5K$conversational scenarios, *(iii)* matches personas to scenarios via semantic similarity, *(iv)* generates $\sim$417K role-play conversations with an LLM where the user speaks as the persona and the assistant behaves as a helpful agent, and *(v)* produces speaker-conditioned user-turn audio (synthetic) from reference recordings to preserve speaker diversity. We evaluate synthetic and human recorded conversations and provide an analysis. We will make the MENASpeechBank available for the community.([https://huggingface.co/datasets/QCRI/MenaSpeechBank](https://huggingface.co/datasets/QCRI/MenaSpeechBank))"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sheikh-ali-etal-2026-menaspeechbank">
<titleInfo>
<title>MENASpeechBank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for SpeechLLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zien</namePart>
<namePart type="family">Sheikh Ali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hunzalah</namePart>
<namePart type="family">Hassan Bhatti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rabindra</namePart>
<namePart type="given">Nath</namePart>
<namePart type="family">Nandi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shammur</namePart>
<namePart type="given">Absar</namePart>
<namePart type="family">Chowdhury</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Firoj</namePart>
<namePart type="family">Alam</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 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Audio large language models (AudioLLMs) enable instruction following over speech and general audio, but progress is limited by the scarcity of *diverse, conversational, and instruction-aligned* speech-text data. This gap is particularly pronounced for persona-grounded and dialectal interactions, where collecting real multi-speaker recordings remains costly and slow. We introduce **MENASpeechBank**, a reference speech bank comprising **\sim18K high-quality utterances** from **124 speakers** spanning multiple MENA countries, covering English, Modern Standard Arabic (MSA), and regional Arabic varieties. We develop a controllable data pipeline that *(i)* constructs persona profiles enriched with World Values Survey (WVS) inspired attributes, *(ii)* defines a taxonomy driven \sim5Kconversational scenarios, *(iii)* matches personas to scenarios via semantic similarity, *(iv)* generates \sim417K role-play conversations with an LLM where the user speaks as the persona and the assistant behaves as a helpful agent, and *(v)* produces speaker-conditioned user-turn audio (synthetic) from reference recordings to preserve speaker diversity. We evaluate synthetic and human recorded conversations and provide an analysis. We will make the MENASpeechBank available for the community.([https://huggingface.co/datasets/QCRI/MenaSpeechBank](https://huggingface.co/datasets/QCRI/MenaSpeechBank))</abstract>
<identifier type="citekey">sheikh-ali-etal-2026-menaspeechbank</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2700/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>61422</start>
<end>61444</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MENASpeechBank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for SpeechLLMs
%A Sheikh Ali, Zien
%A Hassan Bhatti, Hunzalah
%A Nandi, Rabindra Nath
%A Chowdhury, Shammur Absar
%A Alam, Firoj
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F sheikh-ali-etal-2026-menaspeechbank
%X Audio large language models (AudioLLMs) enable instruction following over speech and general audio, but progress is limited by the scarcity of *diverse, conversational, and instruction-aligned* speech-text data. This gap is particularly pronounced for persona-grounded and dialectal interactions, where collecting real multi-speaker recordings remains costly and slow. We introduce **MENASpeechBank**, a reference speech bank comprising **\sim18K high-quality utterances** from **124 speakers** spanning multiple MENA countries, covering English, Modern Standard Arabic (MSA), and regional Arabic varieties. We develop a controllable data pipeline that *(i)* constructs persona profiles enriched with World Values Survey (WVS) inspired attributes, *(ii)* defines a taxonomy driven \sim5Kconversational scenarios, *(iii)* matches personas to scenarios via semantic similarity, *(iv)* generates \sim417K role-play conversations with an LLM where the user speaks as the persona and the assistant behaves as a helpful agent, and *(v)* produces speaker-conditioned user-turn audio (synthetic) from reference recordings to preserve speaker diversity. We evaluate synthetic and human recorded conversations and provide an analysis. We will make the MENASpeechBank available for the community.([https://huggingface.co/datasets/QCRI/MenaSpeechBank](https://huggingface.co/datasets/QCRI/MenaSpeechBank))
%U https://aclanthology.org/2026.emnlp-main.2700/
%P 61422-61444
Markdown (Informal)
[MENASpeechBank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for SpeechLLMs](https://aclanthology.org/2026.emnlp-main.2700/) (Sheikh Ali et al., EMNLP 2026)
ACL