
AI research and products that transform how we interact with technology. Leading foundational models powering ElevenAgents, @ElevenCreative, and ElevenAPI.
Joined January 2022
- Tweets1.9K
- Following11
- Followers213K
- Likes1.6K
Pinned Tweet
Today at ElevenLabs Summit NYC, we shared what we’re building next for agents. Here’s what we announced:
Fine-tuning: In the coming months, enterprises can work with our team to continually fine-tune our Text to Speech and Speech to Text models, as well as open-source LLMs, on their own data.
Interaction Engine: A new system that picks the right LLM for each moment of a conversation. It uses less expensive models for simple requests and frontier models for complex ones like refunds, so you get the right outcome without overspending on tokens.
Memory in ElevenAgents: Agents remember what gets mentioned in passing, so businesses can act on it.
Routines in ElevenAgents: Agents can now run operations before and after conversations, like pre-screening a mortgage applicant and calling them back, or texting about a confusing bill before it turns into a dispute.
ElevenAI: A new AI assistant for work that lets you dictate into any app, takes meeting notes, passes context between agents and people, and provides live assistance during calls.
New on our blog: What it takes for a conversational agent to investigate a denied insurance claim.
When a claim is denied, someone usually has to call the insurer to find out why. A single call can take a medical biller anywhere from 10 minutes to an hour. Every payer has its own phone tree, and resolving the call means getting a representative to explain both what went wrong and how to fix it.
Candid Health has been using ElevenAgents to make these calls. Using the speech and turn-taking models we build in-house, their team tuned the agents’ tone and pacing, down to when they pause. In a pilot of 12,860 calls, the agents found both the reason for the denial and the next step in 59% of cases.
“We deliberately set a high bar for success because getting part of the answer doesn’t really help our billers. We needed the agents to handle the entire call, from understanding why a claim was denied to figuring out what to do next. The fact that they could do that in over half of cases, especially this early, exceeded our expectations.” - Doug Proctor, Co-founder and Chief Operating Officer
elevenlabs.io/blog/candid-he…
ElevenAgents for Healthcare helps teams build conversational agents trusted by patients and clinicians alike
Built on Scribe v3 Live Medical, agents can book appointments, check claims, and answer patient calls around the clock.
Teams at Banner Health, Angle Health, Pfizer, AstraZeneca, and Clover Health have already deployed agents across their workflows.
Scribe v3 Live Medical brings medical transcription to live calls.
It helps voice agents get medications, procedures, and symptoms right, and teams can add up to 10,000 of their own terms, like provider and drug names.
Now in ElevenAgents.
elevenlabs.io/blog/scribe-v3…
. @Kotoba_tech, the leading voice AI lab in Japan, is joining ElevenLabs.
Together, we’ll bring Kotoba’s Japanese voice research to ElevenAgents, so more businesses can serve customers naturally in Japanese.
Welcome to ElevenLabs, @noriyuki_kojima, @jungokasai, and team.
Kotoba is joining ElevenLabs
Most enterprise AI solutions handle every customer interaction the same way. On @SquawkCNBC, @mati shared what’s next for ElevenLabs:
“Customers will be able to fine-tune and train the models with their own data, and extend their advantage based on every interaction.”
Announcing the ElevenLabs Partner Program.
More than 550 certified partners in 44 countries are ready to help you get voice agents into production, from integrating with your existing systems to complex deployments alongside our engineers.
elevenlabs.io/blog/partner-p…
Today we’re opening the ElevenLabs Marketplace.
It brings together agents and applications built on ElevenLabs for healthcare, financial services, insurance, education, retail, and more.
elevenlabs.io/marketplace
“My ultimate goal is to have people stop pressing zero and yelling ‘human’ on the phone and instead say, ‘AI agent.’”
@ashleyekramer joined @MTSlive to talk about why people will soon prefer talking to an AI agent over waiting on hold.
ElevenAgents works for your customers. ElevenAI works for your team.
At ElevenLabs Summit we previewed a new AI assistant for work that lets you dictate into any app, takes meeting notes, passes context between agents and people, and provides live assistance during calls.
Open for early access today.
With Memory, agents can remember. Now they'll be able to follow through, too.
Routines in ElevenAgents handle the work before and after conversations, like pre-screening a mortgage applicant and calling back within minutes, or texting customers to explain a confusing bill before it turns into a dispute.
No one likes to repeat themselves. That’s why we’re bringing Memory to ElevenAgents.
Agents remember what gets mentioned in passing, so businesses can act on it.
At the ElevenLabs Summit, @mati previewed Interaction Engine, a new system that composes and orchestrates the models behind conversational agents.
It uses less expensive models for simple requests and frontier models for complex ones like refunds, so you get the right outcome without overspending on tokens.
Enterprises will soon be able to fine-tune the models behind their voice agents.
Every industry has its own context and language. In the coming months, enterprises can bring their own data and work with our team to fine-tune Text to Speech, Speech to Text, and open-source LLMs.
Each model can be tuned towards the outcomes you measure, such as customer satisfaction or reliability, and will keep improving with each interaction.
A great voice agent has to understand you, sound human, and respond fast.
In the last month, we launched three models: Scribe v3 for transcription accuracy, Eleven v4 for expressive speech, and Eleven Realtime for speed.
Introducing Scribe v3 and Scribe v3 Live: our most accurate Speech to Text models yet.
Built for real-world audio, Scribe excels in situations with extreme background noise, knows pauses from the end of sentences, and keeps track of who said what in a busy meeting.
Scribe v3 has the best transcription and diarization (labeling who said what, when) accuracy of any model we tested.
Scribe v3 Live supports diarization for the first time and makes 43% fewer errors than its predecessor. Turn detection now uses the whole conversation rather than silence alone. A new key term library holds up to 10,000 terms, enough for a full legal or medical vocabulary, and Scribe v3 supports transcript edits in plain English.
In ElevenAgents, @Eneco’s voice agent captured Dutch postal codes correctly on the first try 40% more often.
Scribe v3 Live is available in ElevenAgents today, and both models are available in ElevenAPI. Scribe v3 is also the default in the ElevenLabs app.
Learn more: elevenlabs.io/blog/scribe-v3






