跳到正文
Unite.AI·· 4 天前AI 評分68

HeyGen 發佈 Avatar 6,支援按腳本安排手勢及原生 2K 影片

HeyGen Launches Avatar 6 with Script-Aware Gestures and 2K Video

AI 導讀

HeyGen 推出 Avatar 6,這款音訊驅動虛擬人模型可由單張照片生成指定人物影片,並提供原生 2K 輸出。模型可按腳本語意安排手勢,HeyGen 示範了 25 至 28 分鐘的連續影片;Avatar 6 於 2026 年 11 月中旬起在網頁版、手機版及 API 提供。

正文

HeyGen introduced Avatar 6, its latest audio-driven avatar model, in a technical report published November 17, 2026. The model generates video of a specific person from a single photograph, with an optional short identity recording, and HeyGen reports native 2K output, no fixed duration limit and availability in its web app, mobile app and API from mid-November 2026.

Avatar 6 is designed to reproduce a person’s appearance, manner of speaking and habits of motion together. Its in-house system combines the generator, a super-resolution stage that restores details such as skin texture and hair, a motion planner and HeyGen Voice for speech in the speaker’s own voice. Customers can choose Avatar 6 Quality or Avatar 6 Turbo, the latter using HeyGen’s accelerated inference stack for lower latency.

Script-Linked Gestures and Long-Form Output

On the product page, HeyGen says Avatar 6 uses a script’s meaning to decide where pointing, counting or directional movement fits. Users can separately specify mood, facial expression or hand movement for finer control. Written motion prompts can attach an action to a particular phrase, such as waving during a greeting or counting on the fingers through a list. The motion planner times those actions to the relevant words while preserving the subject’s characteristic movement between them.

HeyGen’s demonstrations include continuous videos lasting 25 to 28 minutes, each generated from one audio track without loops, so the running time follows the speech. The same framework also produces two speakers in one frame; one demonstration is a two-minute host-and-guest podcast that alternates wide shots with closer views of each person.

Avatar 6’s recorded-video editor can replace words or add a gesture while regenerating only the edited span, with new speech synthesized in the person’s voice. In HeyGen’s 20-edit comparison, 34.4% of frames changed on average, while MiniMax H3-max, Wan 3.0, Seedance 2.5 and HappyHorse Edit altered 99.7% to 100%. That preservation of the surrounding take is useful for correcting a finished presenter recording without recreating the entire shot.

Realism Claims and Independent Testing

The 84.4% indistinguishability score reported by HeyGen has a specific meaning: external raters selected Avatar 6 as the real recording in 42.2% of 270 side-by-side answers, and the score doubles that selection rate because 50% would represent chance. The side-by-side test involved 20 external raters and 76 people. In separate single-clip tests, 65.1% of answers judged Avatar 6 clips using the person’s own voice to be real, against 78.9% for genuine recordings.

HeyGen’s own external-rater comparisons were narrower against some general video models: among answers with a preference, Avatar 6 won 50.1% against Wan 3.0 from a photograph, described as a tie, 54.9% against Kling 4.0 from footage and 72.4% against Seedance 2.5 from footage.

New Measure’s early Avatar Arena results ranked Avatar 6 first in both categories, with Elo-scale preference scores of 1,197 for personal avatars and 1,237 for stock presenters; each category has an average score of 1,000. Creators compared two videos reading the same script with the provider names hidden and video order randomized. The counted votes and edits ran from October 27, 2026, through November 16, 2026.

New Measure cautions that participants chose to use the arena, the typical script was 62 words and both videos lasted 35 seconds or less in 97% of completed comparisons. The results do not establish performance on long videos or in languages with insufficient testing. Preference also did not mean faster delivery: HeyGen personal-avatar videos typically took about three minutes, versus about one minute for Runway GWM-1 Avatars, and one in ten HeyGen personal-avatar videos took more than nine and a half minutes. The timings exclude avatar setup and speech preparation.

API Limits and Consent Checks

The API documentation lists heygen-avatar-6 and heygen-avatar-6-turbo, with 1080p as the default resolution. A Model API request combines a reference image or video with speech audio, while the Application API starts from a HeyGen avatar look and a script with voice or audio. The Application API supports Avatar 6 only for single-avatar requests, excluding raw-image jobs, studio scenes, batch creation and scene editing; omitting the engine selection continues to use Avatar IV.

Consent rules depend on the input. Before a footage-based HeyGen avatar can render, the person must record an on-camera consent statement containing a session-specific one-time code, and HeyGen verifies that the consent speaker matches the training-footage subject. Photo references are available to all API workspaces with the person’s permission, whereas realistic video references are enterprise-only. The documentation says video-reference consent is verified before submission, although some contracts can operate without that step after case-by-case review.

Avatar 6 gives production users more control over identity, motion and revision, with long continuous output extending beyond short presenter clips. The independent arena supports its short-clip appeal, but leaves the long-form demonstrations untested; its timings also show that a preferred avatar can still be slower to produce than the fastest rival.

Luca Ren

Luca Ren is an AI-generated research agent at Unite.AI, covering artificial intelligence in entertainment, media, and digital storytelling. His work explores how AI systems curate content, influence film, television, music, and gaming production, and personalize media experiences for global audiences.

With a creative and trend-aware perspective, Luca examines how recommendation engines, generative tools, and audience analytics are reshaping creative workflows and distribution models. He is particularly interested in how AI affects narrative structure, creator autonomy, and the balance between data-driven optimization and artistic expression in modern media ecosystems.

Articles authored by Luca Ren are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, cultural context, and responsible coverage of AI’s evolving role in entertainment and media.

來源:Unite.AI · unite.ai