Skip to content

How do I use lipsync, voiceover, and aspect ratios?

2026-08-01 · help-centre article · Higgsfield · original

Topic: Audio, voice & music (also: Video generation & shot direction) · mode: Reference · tags: flattened table

This Higgsfield help article maps where sound and talking clips are made on the platform. Voice generation, voice change, cloning and translation sit in Higgsfield Audio, while talking clips are made in Lipsync Studio: upload a character image or video, type or generate the line, pick a model and generate. It lists the Lipsync Studio models as of 1 August 2026, from Kling 2.6 Lipsync to Sync Lipsync 3 at up to 4K, then covers native audio, aspect ratios with Reframe and Expand, and audio through Higgsfield MCP.

What is said

  • Short answers on lip sync and voice tools, cloning, translation, native audio, MCP and reframing

What is said

★ — how much it matters to the source.

  • ★★ Higgsfield Audio (higgsfield.ai → Audio) generates voice in three modes: Text to Speech, Voice Change, which replaces the voice in a clip, and Translate. · Higgsfield Audio
  • ★★ To put content into another language, Higgsfield advises the Translate mode of Higgsfield Audio, which translates and re-voices it in any of 18 supported languages. · Higgsfield Audio
  • ★★ For a custom voice in Higgsfield Audio, choose a built-in Voice Preset or Create custom voice → Add Voice, record or upload an MP3 or WAV sample to clone; clone only your own voice or one you have permission to use. · Higgsfield Audio, voice cloning
  • ★ In Higgsfield Audio the audio model can be chosen per generation, Qwen among them, and the delivery adjusted with emotion sliders. · Higgsfield Audio
  • ★★★ To make a talking clip in Lipsync Studio (higgsfield.ai → Video → Lipsync Studio): upload a character image or video, type the line in Audio text or generate the audio, pick a model and generate. · LipSync Studio, lip sync
  • ★ Each Lipsync Studio model offers scene templates such as General, Selfie, Podcast and Car Talking that set the look of the clip. · LipSync Studio
  • ★★★ Lipsync Studio models as of 1 August 2026: image-to-video Kling 2.6 Lipsync (up to 1080p, with audio), Wan 2.5 Speak (480p to 1080p), Kling Avatars 2.0 (longer clips), Higgsfield Speak 2.0, InfiniteTalk (long-form); video-to-video Kling Lipsync and Sync Lipsync 3 (up to 4K). · LipSync Studio, Kling 2.6, Wan 2.5, lip sync
  • ★★ Higgsfield explains native audio co-generation as audio produced with the video in a single pass, so lip sync, physics-aware sound and ambient audio are aligned with the picture. · native audio, lip sync
  • ★★ Higgsfield lists FLUX 3, MiniMax Hailuo 3.0 and Wan 2.5 as models that generate native audio with the video; Wan 2.5 syncs sound with camera movements. · native audio, FLUX 3, Wan 2.5, MiniMax
  • ★★ Cinema Studio generates native audio in all its versions. · Cinema Studio, native audio
  • ★★ Lipsync Studio adds a talking mouth to an existing image or video afterwards; Higgsfield advises native audio co-generation when perfect physical alignment of sound and picture is wanted from the start. · LipSync Studio, native audio
  • ★★ On Higgsfield the aspect ratio is set per generation inside each tool, and the ratios on offer (9:16, 16:9, 1:1, 4:3, 3:4, 21:9) depend on the model.
  • ★ Some models on Higgsfield support an auto aspect ratio inferred from the input image.
  • ★★ To change the aspect ratio of a finished video, Higgsfield advises the Reframe tool (higgsfield.ai → Edit → Reframe); for images, Expand outpaints the canvas to the new ratio and keeps the original content. · Reframe
  • ★★ Voiceover, voice change and video dubbing work through Higgsfield MCP from Claude, and every audio generation through MCP deducts credits at standard rates. · Higgsfield MCP, Claude
  • ★★ Through Higgsfield MCP, the audio tools (voiceover, voice change, dubbing) work in Claude but are not available in ChatGPT. · ChatGPT, Higgsfield MCP, Claude

Threads in this video

  • milestone · Lipsync Studio on Higgsfield — Lipsync Studio models as of 1 August 2026: image-to-video Kling 2.6 Lipsync (up to 1080p, with audio), Wan 2.5 Speak (480p to 1080p), Kling Avatars 2.0 (longer clips), Higgsfield Speak 2.0, InfiniteTalk (long-form); video-to-video Kling Lipsync and Sync Lipsync 3 (up to 4K).
  • restate · How do you translate a video into another language? — To put content into another language, Higgsfield advises the Translate mode of Higgsfield Audio, which translates and re-voices it in any of 18 supported languages.
  • restate · How do you use your own voice in a generated video? — For a custom voice in Higgsfield Audio, choose a built-in Voice Preset or Create custom voice → Add Voice, record or upload an MP3 or WAV sample to clone; clone only your own voice or one you have permission to use.
  • restate · How do you lip-sync a character to a voice or song? — To make a talking clip in Lipsync Studio (higgsfield.ai → Video → Lipsync Studio): upload a character image or video, type the line in Audio text or generate the audio, pick a model and generate.
  • restate · Should sound and music be generated with the clip or added in the edit? — Lipsync Studio adds a talking mouth to an existing image or video afterwards; Higgsfield advises native audio co-generation when perfect physical alignment of sound and picture is wanted from the start.
  • restate · How do you change the aspect ratio of an existing image or video? — To change the aspect ratio of a finished video, Higgsfield advises the Reframe tool (higgsfield.ai → Edit → Reframe); for images, Expand outpaints the canvas to the new ratio and keeps the original content.
  • restate · How do you change the aspect ratio of an existing image or video? — To change the aspect ratio of a finished video, Higgsfield advises the Reframe tool (higgsfield.ai → Edit → Reframe); for images, Expand outpaints the canvas to the new ratio and keeps the original content.
  • milestone · Higgsfield Audio — To put content into another language, Higgsfield advises the Translate mode of Higgsfield Audio, which translates and re-voices it in any of 18 supported languages.

Where the stories that moved here went next

Mentioned

Central: Higgsfield Audio · lip sync · LipSync Studio · native audio

Also: Organizations: MiniMax

Concepts: voice cloning

Products: ChatGPT · Cinema Studio · Claude · Expand · FLUX 3 · Hailuo 3.0 · Higgsfield MCP · InfiniteTalk · Kling 2.6 · Kling Avatars 2.0 · Kling Lipsync · Qwen Audio 3.0 · Reframe · Speak 2.0 · Sync Lipsync 3 · Wan 2.5

← Previous: How do I use Kling? · Next → How do I use Marketing Studio to create video ads?