# Seed Speech

By Steven Van · 2026-04-13

ByteDance's text-to-speech and speech recognition engine covers 10 languages and sits under Audio in Higgsfield.

Higgsfield now includes [Seed Speech](<https://higgsfield.ai/creator-hub/changelog#seed-speech>), ByteDance's engine that handles text-to-speech and speech recognition in one tool. It is aimed at dubbing, voiceover, subtitles and AI assistants.

## Text-to-speech

The TTS side generates voice with emotion and tone controlled through text prompts. It supports 10 languages.

## Speech recognition

The ASR side transcribes speech with automatic punctuation. It also detects speakers and emotion.

## Where to find it

In [Higgsfield](<https://creatorstoolbox.com/tools/higgsfield>), open Audio, then Seed Speech.

![Higgsfield](<https://higgsfield.ai/icon.png>)

Higgsfield

AI video and image generation platform — create videos, edit with motion control, swap faces, and collaborate in real time.

[View Higgsfield →](<https://creatorstoolbox.com/tools/higgsfield>)

Sources

[Original source](<https://higgsfield.ai/creator-hub/changelog#seed-speech>)

[Read Seed Speech on Creators Toolbox](<https://creatorstoolbox.com/blog/higgsfield-seed-speech>)

---
Canonical source: https://creatorstoolbox.com/blog/higgsfield-seed-speech