Humans are tribal creatures. We love our local sports teams, we build communities with the people who share our neighborhoods, and we base our identities around the shared connections we have with others. Accents, slang, and regional dialects are key parts of that cultural vernacular. Most people can remember the first time they traveled away from their community and used a piece of local slang that drew vacant looks from that community’s “locals,” and how refreshing it felt to return to a place where everyone sounded just like they did. There’s an inherent sense of trust and community that is built when someone sounds like you.
I can hear you asking the question now: what do regional accents have to do with multifamily? It turns out, quite a bit. Hundreds of leasing and marketing teams across the nation have deployed EliseAI’s VoiceAI product to decrease missed call rates and convert leads to tours to leases around the clock. The industry has made their feedback clear: VoiceAI is so lifelike that a large percentage of callers are unable to distinguish it from a live human, converting leases-to-leads at the same (if not better) rates than many live agents. It has achieved those numbers without the impact of local accents, as the VoiceAI product has exclusively relied on generic, unaccented voices to date.
With that in mind, the EliseAI Research and Engineering teams asked a simple question: what would the impact of switching from generic AI voice agents to localized AI voice agents do for leasing conversion rates?
.svg)
The Experiment
In late November 2025, EliseAI worked with Elevenlabs and Cartesia, two of the leading creators of AI voice models, to collect a series of high level voices with specific regional accents curated from their robust voice library. EliseAI worked closely with each model company on the right settings and configuration for the voices, iterating on things like speed of the voice and stability to ensure optimal performance. The EliseAI team subsequently chose a randomized subset of customer properties to move away from a standardized AI voice rotation and instead implemented "Voice Localization,” assigning voices with regional tones and inflections to properties in matching geographic locations (e.g., Southern properties received Southern voices). The team continually iterated on the voices based on how residents and prospects respond to them, measuring caller engagement and tour conversions as well as tracking hangups and transfers.
The underlying hypothesis was that, given existing research showing consumers are more likely to transact with people who sound like them, that switching to localized voices would positively impact lead-to-tour conversion rates. The overall results proved that hypothesis, and more.
The Outcomes
Across the subset of properties selected for the experiment, those with localized AI voice agents outperformed the non-localized agents significantly across a variety of statistical factors. Lead-to-tour conversion rates, the primary metric for the experiment, increased by 5.28% on localized voice communities, definitively indicating that prospective renters are in fact more likely to rent an apartment or home from a leasing agent (or in this case, AI agent) who has a regional accent reflecting their unique locality. Taking into account the trend data, the one-sided p-value for this experiment is p ≈ 0.0335, which indicates a statistically significant sample size rather than just a happy coincidence.
It’s easy to understand why this experiment was a success. Leasing a new home is a large financial commitment, and consumers inherently trust local voices that could be perceived as more trustworthy or well informed about a certain neighborhood or locality. There’s also a mimicry effect at play here—just as waitresses who mimic customers get 68% higher tips (van Baaren et al.), our VoiceAI agents that "mimic" the local dialect see higher conversion rates. When all taken together, the combination of perceived local knowhow and voice “sameness” has driven significantly improved leasing outcomes for housing operators leveraging the VoiceAI product.
There were also other key underlying data points that indicated that the localized voice shift was driving improved conversion rates and conversation quality. There was an 7.8% increase in overall caller engagement with the VoiceAI product, as customers conducted richer, fuller conversations with our AI agents. Median words spoken on a per caller basis increased by 15.8%, as callers had deeper, more comprehensive conversations with the AI. Finally, hang-ups occurring under 10s dropped by 12.1%, as fewer customers hung up either immediately or early in the call.
When taken in summary, the results unequivocally show that switching to localized voices increased VoiceAI conversion rates, improved average conversation quality, and helped people get what they need with less friction… demonstrating that the “local touch” clearly makes a difference when it comes to leasing apartments.
What This Means for VoiceAI Customers Moving Forward
Unlike other providers of AI voice technology that rely on generic voices and a “one-size-fits-all” approach to product development, the EliseAI Research and Product team continues to iterate and improve our solutions based on both customer feedback and real-world experimentation like the project conducted here. Instead of trying to create the perfect, neutral “corporate” voice, we continue to push the boundaries of what is possible with VoiceAI technology, leveraging existing academic research about demonstrated sociological trends to ensure our products enable our housing and healthcare customers to drive the best outcomes for their businesses.
Moving forward, EliseAI is planning on continuing to test newer, more realistic voices in partnership with voice-AI model providers, ensuring our catalog steadily improves in quality and naturalness. Our goal is to give customers clear, data-driven insights on which voices best fit their residents while expanding to more granular regions that reflect true linguistic and cultural boundaries, all the while improving asset performance.
Increase Conversion Rates with VoiceAI
The combination of VoiceAI’s new localization capabilities, the ability of the solution to speak over 50 languages (while converting more leads-to-tours in those languages), and industry-best response times and conversational quality make it clear that the VoiceAI product is the foremost agentic voice AI product serving the housing and healthcare markets. Interested in learning more about how VoiceAI can help make sure you never miss the chance to turn an inbound call to a new lease? Get in touch with us by filling out the form below to learn more.