Infer.so creates the world’s first multimodal AI voice bot

Infer.so is a California-based all-in-one AI solution for sales call handling, management, and training. Its AI voice bots automate phone conversations and execute tasks like humans, with use cases in industries like insurance, e-commerce, hospitality, fintech, and more.

The bots—which have infinite memory and personalization capabilities—use machine learning to understand conversational themes. This allows these AI-powered bots to respond to users just like humans would.

Having understood what users are asking and formulated a response, Infer.so then uses ElevenLabs to power its responses.

Selecting ElevenLabs 

Infer.so needed a platform that could generate lifelike voices to suit their range of use cases.

"[Our] initial discovery of ElevenLabs came from browsing Hacker News and conducting thorough research on various voice cloning and synthesis services available at the time", says the Infer.so team.

After examining a broad range of options, they were struck by ElevenLabs’ ability to generate authentic-sounding voices.

"ElevenLabs' voices stood out as the most realistic and user-friendly."

However, it didn’t want to sink time and money into integration—a common hiccup for start-ups onboarding new software. So, it was delighted to discover that the ElevenLabs integration process was slick and efficient.

"The entire procedure required minimal effort and time, which allowed us to quickly incorporate the technology into our projects without any significant hurdles."

Of course, integration is just the first piece of the puzzle. Infer.so prioritizes quality above all else. Its bots are used in a wide range of business environments—so it must ensure they're as natural and human-like as possible.

ElevenLabs has consistently stood head and shoulders above the rest of the market in this regard.

"Despite the emergence of new solutions from larger incumbents over time, we remained loyal to ElevenLabs, primarily due to the superior voice quality offered."

Infer.so now uses ElevenLabs to synthesize speech in real-time for their voice agents. This is no mean feat, requiring Infer.so to deliver in terms of both quality and response generation times.

The partnership with ElevenLabs has delivered on both these fronts.

“ElevenLabs has impressively managed to maintain high-quality voice outputs while significantly minimizing latency, ensuring that our voice agents can respond quickly and naturally”. 

In turn, this has led to users feeling better supported by chatbots. "This capability has been instrumental in enhancing the interaction experience between our agents and users, providing seamless and realistic conversations."

ElevenLabs has played a key role in supporting Infer.so's growth, and will continue to do so moving forward.

"ElevenLabs has become our go-to for real-time voice synthesis, ensuring our voice agents deliver quick, natural responses with unmatched quality." 

Trial Infer.so’s multimodal voice bot, supported by ElevenLabs.