Skip to content

Changelog: Audio. Redesigned, with a new model

2026-08-10 · changelog: feature release · Higgsfield · original

Topic: Audio, voice & music · mode: Release note

Higgsfield's changelog (10 August 2026) says Audio has been redesigned, with a new model, Qwen, and advanced settings for emotion, format and pitch; its three functions are Text to Speech, Voice Change and Translate, the last covering 18 languages with lip-sync.

What is said

  • Audio. Redesigned, with a new model

What is said

★ — how much it matters to the source.

  • ★★★ Higgsfield has redesigned Higgsfield Audio and added a new model, Qwen, along with advanced settings for its audio models: control over emotion, format, pitch and more. · Higgsfield Audio
  • ★★ Text to Speech in Higgsfield Audio generates speech from text with a preset voice, or with a voice the user uploads or records. · Higgsfield Audio
  • ★★ Voice Change in Higgsfield Audio swaps the voice in an uploaded video while keeping the original delivery. · Higgsfield Audio
  • ★★ Translate in Higgsfield Audio translates a video into another language and lip-syncs it, choosing from a library of 18 languages. · Higgsfield Audio, lip sync
  • ★ To start in Higgsfield Audio, open Audio and pick Text to Speech, Voice Change or Translate. · Higgsfield Audio

Threads in this video

  • milestone · Higgsfield Audio — Higgsfield has redesigned Higgsfield Audio and added a new model, Qwen, along with advanced settings for its audio models: control over emotion, format, pitch and more.

Where the stories that moved here went next

Mentioned

Central: Higgsfield Audio

Also: lip sync · Qwen Audio 3.0

← Previous: Add AI VFX to Real Footage: Set your head on fire · Next → Blockbuster 4K: The AI Filmmaking Pipeline: Scene 1: The Sea Battle and Escape Overboard