---
type: "post"
id: "eleven-v4-our-most-expressive-text-to-speech-bff1c3"
slug: "eleven-v4-our-most-expressive-text-to-speech-bff1c3"
title: "Eleven v4: Our most expressive text-to-speech AI model yet"
description: "Thesis: Expressive speech is being positioned as a latency-compatible capability for voice agents, rather than a quality reserved for slower generation."
retrieval_nugget: "Thesis: Expressive speech is being positioned as a latency-compatible capability for voice agents, rather than a quality reserved for slower generation."
published_at: "2026-10-05"
updated_at: "2026-10-05"
record_date: "2026-10-05"
date_kind: "published_at"
topics: []
source_urls: ["https://elevenlabs.io/blog/eleven-v4"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:eleven-v4-our-most-expressive-text-to-speech-bff1c3"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:1334c1bc092bf343bc5142e9bab1e73937b77adefc7781efd219b54a1afef2dc","reviewed_at":"2026-10-05T20:05:19.659Z","source_evidence_count":1,"verified_claim_count":2,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":2,"implication_count":2,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Expressive speech is being positioned as a latency-compatible capability for voice agents, rather than a quality reserved for slower generation.","observed_facts":[{"text":"ElevenLabs announced Eleven v4 and the low-latency v4 Turbo variant.","source_urls":["https://elevenlabs.io/blog/eleven-v4"]},{"text":"The company says both support more than 90 languages and reports about 150 ms median time to first speech for Turbo.","source_urls":["https://elevenlabs.io/blog/eleven-v4"]}],"mechanism":"The announcement describes a model intended to use tone, pacing, emotion, and conversational context, while Turbo targets faster response. If these capabilities hold in deployment, agent builders may be able to combine quicker turn-taking with speech that sounds less mechanically uniform.","why_now":"The launch directly addresses a familiar product tension in voice interfaces: fast responses can sound flat, while more expressive generation can add delay.","implications":["Developers may evaluate whether expressive output improves user experience enough to justify integration and tuning work.","The model’s advertised conversational dynamics could also matter for multi-speaker applications, but product claims alone do not establish reliable performance in those settings."],"evidence_boundary":"The available evidence is ElevenLabs’ own product announcement. Its latency figures, comparative rankings, and listener-preference results are company-reported and are not independently verified here.","watch_conditions":["Independent latency tests across realistic agent stacks would show whether Turbo’s claimed speed and expressive output survive deployment conditions."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this captures the specific release and the speed-versus-expressiveness question it raises."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/eleven-v4-our-most-expressive-text-to-speech-bff1c3/","markdown":"https://newruntime.com/posts/eleven-v4-our-most-expressive-text-to-speech-bff1c3.md","json":"https://newruntime.com/posts/eleven-v4-our-most-expressive-text-to-speech-bff1c3.json"}
---

# Eleven v4: Our most expressive text-to-speech AI model yet

## Retrieval answer

Thesis: Expressive speech is being positioned as a latency-compatible capability for voice agents, rather than a quality reserved for slower generation.

Thesis: Expressive speech is being positioned as a latency-compatible capability for voice agents, rather than a quality reserved for slower generation.

Observed facts: ElevenLabs announced Eleven v4 and the low-latency v4 Turbo variant. The company says both support more than 90 languages and reports about 150 ms median time to first speech for Turbo.

Mechanism: The announcement describes a model intended to use tone, pacing, emotion, and conversational context, while Turbo targets faster response. If these capabilities hold in deployment, agent builders may be able to combine quicker turn-taking with speech that sounds less mechanically uniform.

Why now: The launch directly addresses a familiar product tension in voice interfaces: fast responses can sound flat, while more expressive generation can add delay.

Implications: Developers may evaluate whether expressive output improves user experience enough to justify integration and tuning work. The model’s advertised conversational dynamics could also matter for multi-speaker applications, but product claims alone do not establish reliable performance in those settings.

Evidence boundary: The available evidence is ElevenLabs’ own product announcement. Its latency figures, comparative rankings, and listener-preference results are company-reported and are not independently verified here.

Watch conditions: Independent latency tests across realistic agent stacks would show whether Turbo’s claimed speed and expressive output survive deployment conditions.

The announcement makes this a distinct analytical record because no prior New Runtime record was supplied; it captures the specific release and the speed-versus-expressiveness question it raises.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)