--- title: Eleven v4 Turbo Gets Voice Agents to Full Duplex description: "ElevenLabs shipped Eleven v4 Turbo today: 150ms to first speech, streaming both ways, a turn-taking model around it. The full duplex setup I wrote about in August." canonical: https://lukeocodes.dev/eleven-v4-turbo-full-duplex type: article locale: en_GB published: 2026-09-28T00:00:00.000Z modified: 2026-09-28T00:00:00.000Z authors: - Luke Oliff section: Voice AI tags: - Voice AI - Industry - Latency - elevenlabs - eleven v4 - text to speech - full duplex - voice agents - streaming - model release feed: https://lukeocodes.dev/rss.xml search: https://lukeocodes.dev/search.json?q={query} schema: - "@context": https://schema.org "@graph": - "@type": WebSite "@id": https://lukeocodes.dev/#website url: https://lukeocodes.dev name: Luke Oliff alternateName: lukeocodes inLanguage: en publisher: "@id": https://lukeocodes.dev/authors/luke#person description: Personal blog and writing by Luke Oliff — developer relations, voice AI, developer experience. potentialAction: - "@type": SearchAction target: "@type": EntryPoint urlTemplate: https://lukeocodes.dev/blog/search?q={search_term_string} query-input: required name=search_term_string - "@type": Person "@id": https://lukeocodes.dev/authors/luke#person name: Luke Oliff url: https://lukeocodes.dev/authors/luke description: Engineer and developer relations leader. Writing about developer relations, AI, and the voice industry, from the UK. mainEntityOfPage: https://lukeocodes.dev/authors/luke givenName: Luke familyName: Oliff image: https://lukeocodes.dev/authors/luke.jpg jobTitle: Engineer and developer relations leader sameAs: - https://www.linkedin.com/in/lukeocodes - https://github.com/lukeocodes - https://bsky.app/profile/lukeocodes.dev - https://x.com/lukeocodes - https://hashnode.com/@lukeocodes - https://dev.to/lukeocodes knowsAbout: - Developer Relations - Developer Experience - Voice AI - Text-to-Speech - Answer Engine Optimisation - Search Engine Optimisation - Open Source - Developer Tooling - API Design - Speech-to-Text - Voice Agents - Community Building knowsLanguage: - English nationality: British alumniOf: - "@type": Organization name: The Open University - "@context": https://schema.org "@graph": - "@type": Person "@id": https://lukeocodes.dev/authors/luke#person name: Luke Oliff url: https://lukeocodes.dev/authors/luke description: Engineer and developer relations leader. Writing about developer relations, AI, and the voice industry, from the UK. mainEntityOfPage: https://lukeocodes.dev/authors/luke givenName: Luke familyName: Oliff image: https://lukeocodes.dev/authors/luke.jpg jobTitle: Engineer and developer relations leader sameAs: - https://www.linkedin.com/in/lukeocodes - https://github.com/lukeocodes - https://bsky.app/profile/lukeocodes.dev - https://x.com/lukeocodes - https://hashnode.com/@lukeocodes - https://dev.to/lukeocodes knowsAbout: - Developer Relations - Developer Experience - Voice AI - Text-to-Speech - Answer Engine Optimisation - Search Engine Optimisation - Open Source - Developer Tooling - API Design - Speech-to-Text - Voice Agents - Community Building knowsLanguage: - English nationality: British alumniOf: - "@type": Organization name: The Open University - "@type": BlogPosting "@id": https://lukeocodes.dev/eleven-v4-turbo-full-duplex#article headline: Eleven v4 Turbo Gets Voice Agents to Full Duplex alternativeHeadline: Eleven v4 Turbo Gets Voice Agents to Full Duplex description: "ElevenLabs shipped Eleven v4 Turbo today: 150ms to first speech, streaming both ways, a turn-taking model around it. The full duplex setup I wrote about in August." datePublished: 2026-09-28T00:00:00.000Z dateModified: 2026-09-28T00:00:00.000Z url: https://lukeocodes.dev/eleven-v4-turbo-full-duplex mainEntityOfPage: https://lukeocodes.dev/eleven-v4-turbo-full-duplex inLanguage: en author: - "@id": https://lukeocodes.dev/authors/luke#person isPartOf: "@id": https://lukeocodes.dev/#website keywords: Voice AI, Industry, Latency, elevenlabs, eleven v4, text to speech, full duplex, voice agents, streaming, model release articleSection: Voice AI about: - "@type": Organization name: ElevenLabs url: https://elevenlabs.io - "@type": Product name: Eleven v4 - "@type": Organization name: Artificial Analysis - "@type": Thing name: Voice AI abstract: ElevenLabs released Eleven v4 and v4 Turbo on 28 September 2026. Turbo takes text while your LLM is still writing and sends audio back about 150 milliseconds later. In ElevenAgents a turn-taking model handles interruptions, so you get an agent you can talk over. The full duplex setup from my August post. speakable: "@type": SpeakableSpecification cssSelector: - .post-tldr - h1 image: https://lukeocodes.dev/og.png?h=Eleven+v4+Turbo+Gets+Voice+Agents+to+Full+Duplex&d=ElevenLabs+shipped+Eleven+v4+Turbo+today%3A+150ms+to+first+speech%2C+streaming+both+ways%2C+a+turn-taking+model+around+it.+The+full+duplex+setup+I+wrote+about+in+August.&e=Writing wordCount: 909 timeRequired: PT5M publisher: "@id": https://lukeocodes.dev/authors/luke#person - "@type": BreadcrumbList itemListElement: - "@type": ListItem position: 1 name: Home item: https://lukeocodes.dev - "@type": ListItem position: 2 name: Writing item: https://lukeocodes.dev/blog - "@type": ListItem position: 3 name: Eleven v4 Turbo Gets Voice Agents to Full Duplex item: https://lukeocodes.dev/eleven-v4-turbo-full-duplex - "@type": FAQPage "@id": https://lukeocodes.dev/eleven-v4-turbo-full-duplex#faq mainEntity: - "@type": Question name: Is Eleven v4 Turbo full duplex? acceptedAnswer: "@type": Answer text: "ElevenLabs does not use the phrase. Eleven v4 Turbo is the speaking part: text goes in while audio comes out, with about 150 milliseconds to first speech. ElevenAgents joins it to speech-to-text, your LLM (the text model that writes the reply) and a turn-taking model that decides when the agent stops and when it carries on. Full duplex is my name for that whole setup: an agent you can talk over." - "@type": Question name: What is the difference between Eleven v4 and Eleven v4 Turbo? acceptedAnswer: "@type": Answer text: Eleven v4 is tuned for produced content like audiobooks and video, where quality matters most. Eleven v4 Turbo is the fast version for voice agents, with about 100 milliseconds of processing time and about 150 milliseconds before you hear the first sound. Both have the same range of expression, the same 90 plus languages and both support Professional Voice Clones. - "@type": Question name: How fast is Eleven v4 Turbo? acceptedAnswer: "@type": Answer text: ElevenLabs says the median time to first speech is about 150 milliseconds and median processing time is about 100 milliseconds. Median means the middle result, so half of requests were faster and half slower. Its own comparison puts Cartesia Sonic 3.6 at 262 milliseconds and OpenAI's GPT-4o mini TTS at 814 milliseconds. No independent measurement has been published yet. - "@type": Question name: What else changed in Eleven v4? acceptedAnswer: "@type": Answer text: Language support went from 70 in v3 to more than 90, according to TechCrunch. Instant Voice Clones now need 10 seconds of audio. Professional Voice Clones are back after being missing from v3. A single generation can be up to 10,000 characters. Both models are in ElevenAgents, ElevenCreative and the API, and the free plan gives 10,000 credits a month, about 10 minutes of audio. --- [Luke Oliff.](/) - [Blog](/blog) - [CV](/cv) - [Uses](/uses) - [User manual](/user-manual) - [Living with ADHD](/living-and-working-with-adhd) - [RSS](/rss.xml) [Luke Oliff · Writing](/blog) # Eleven v4 Turbo Gets Voice Agents to Full Duplex Sep 2026 · [Voice AI](/blog/category/voice-ai) · 5 min read · [Luke Oliff](/authors/luke) **TL;DR:** ElevenLabs released Eleven v4 and v4 Turbo on 28 September 2026. Turbo takes text while your LLM is still writing and sends audio back about 150 milliseconds later. In ElevenAgents a turn-taking model handles interruptions, so you get an agent you can talk over. The full duplex setup from my August post. ElevenLabs released Eleven v4 and Eleven v4 Turbo on 28 September 2026. Turbo takes text while your LLM (the text model that writes the reply) is still writing it, and starts sending audio back about 150 milliseconds later. Inside ElevenAgents, a turn-taking model decides when the agent should stop and when it should keep going, so you get an agent you can talk over. ElevenLabs doesn't call that full duplex. I do, because it's the setup [my August whitepaper](https://lukeocodes.dev/full-duplex-voice-ai-new-architecture) asked for. That post listed three problems with how voice agents are built. This one goes through them again with today's release in hand, so it's a follow on. ## What ElevenLabs shipped Two models came out together. [Eleven v4](https://elevenlabs.io/blog/eleven-v4) is for produced content, the audiobook and video work where quality matters more than speed. Eleven v4 Turbo is the fast version for agents. ElevenLabs says both are built on "an entirely new architecture" (a new way of building the model), and the [product page](https://elevenlabs.io/v4) describes something it calls context stitching, which keeps the same voice steady across a whole script. The numbers from the product page: - More than 90 languages. - Instant Voice Clones from 10 seconds of audio. - Professional Voice Clones are back. They weren't supported in v3. - One generation can be up to 10,000 characters. - Free plan: 10,000 credits a month, about 10 minutes of audio. Premium is $6 a month. The blog post adds the test results. ElevenLabs says Artificial Analysis, an outside site that ranks AI models, puts v4 first for September 2026. It says about 75% of listeners picked v4 in blind tests, where the listener doesn't know which model they're hearing, and that v4 won 65% to 81% of head-to-head tests against Cartesia Sonic 3.6, Inworld TTS-2 and two of Google's Gemini voices. Nobody outside ElevenLabs has published a measurement yet. [TechCrunch](https://techcrunch.com/2026/09/28/elevenlabs-new-v4-speech-model-supports-more-expression-control-and-90-languages/) has the business side. Language support was 70 in v3. Revenue run rate, meaning this year's pace rather than money already in, is past $600 million, up from about $330 million at the start of the year, and more than 55% of it comes from big companies. ## What streaming both ways gives you The old way to connect text-to-speech to an agent was to wait for the LLM to finish a sentence, send the sentence, then wait for the audio. Every sentence waited for the full trip out and back. With v4 Turbo you push words as the LLM produces them and audio comes back before the sentence is finished. ElevenLabs' words: "Bidirectional streaming, built for agent loops." Latency, the wait before you hear anything, is about 150 milliseconds to the first sound. The model's own processing time is about 100 milliseconds. Both are medians, the middle result with half faster and half slower. ElevenLabs' comparison table puts Cartesia Sonic 3.6 at 262 milliseconds and OpenAI's GPT-4o mini TTS at 814. ## Is Eleven v4 Turbo full duplex? The three problems from August The whitepaper described a voice agent as a chain of three parts, each waiting for the one before it to finish: speech-to-text, then the LLM, then text-to-speech. I called that chain the cascade. It broke the trouble with it into three problems. The first was latency adding up at every step. Speech-to-text waits for enough audio, the LLM waits for the full transcript, and text-to-speech waits for the full text. v4 Turbo removes the last wait, and streaming both ways means it no longer waits for a full sentence either. The second was that the model knows what it said but not what the user heard. The bigger the gap between what has been generated and what has been played, the more there is to sort out after an interruption. At 150 milliseconds to first speech, with audio going out as it's made, that gap is now tiny. Interruptions are the third problem, and the one I care about most. People make noises while they listen: "mm", "yeah", "uh-huh", a laugh, a cough. On a call with a person those noises mean carry on, but a plain voice detector hears them as the user talking and stops the agent mid-sentence. [ElevenAgents](https://elevenlabs.io/docs/agents-platform/overview) is built from four parts: a speech-to-text model, your LLM, a text-to-speech model (v4 Turbo now, according to the v4 blog post) and what the docs call "a proprietary turn-taking model that handles conversation timing". Proprietary means their own, not public. A model whose only job is timing should be able to learn that a sound is not always a turn, and not always a change of subject. The docs don't say what it's trained on, and they list the interruption and turn-taking settings you can change. ## What to do with it today If you build voice agents, use Turbo for the speaking part, then make sure the rest of the chain, speech-to-text and the LLM, adds no more than another 150 milliseconds. Then measure the three things the whitepaper asked for: the time from the user going quiet to the agent starting to speak, the time it takes to recover from an interruption, and how often the agent carries on talking after the user has started. If you're on ElevenAgents, the turn-taking settings are where you change the last two. One more thing from the agents docs: they still say the text-to-speech there covers 70 plus languages, while the v4 pages say 90 plus. Expect that page to catch up. [← Newer: DevRel This Week, 28 September: Hacktoberfest Drops the Pull Request](/devrel-this-week-28-september-2026) [All posts](/blog) → ## Search this site `GET https://lukeocodes.dev/search.json?q=` returns matching posts as JSON, each with its page URL and its markdown URL. Add `&limit=<1-50>` for more than 10. How it works: [/blog/search](https://lukeocodes.dev/blog/search.md). ```sh curl "https://lukeocodes.dev/search.json?q=developer+relations" ``` --- © 2026 Luke Oliff. Posts live at the root of this site. - [GitHub](https://github.com/lukeocodes) - [LinkedIn](https://www.linkedin.com/in/lukeocodes) - [X](https://x.com/lukeocodes) - [Bluesky](https://bsky.app/profile/lukeocodes.dev) - [Email](mailto:luke@lukeoliff.com) - [RSS](/rss.xml) ## Cookies This site sets one analytics cookie, from Google Analytics, and only in a browser that accepts the cookie banner. A client that does not run scripts sets no cookies, so there is nothing to accept or decline here. [Cookie policy](/cookies)