Higgsfield TTS API

Text to Speech

Text-to-speech with customizable voices. Adjustable speed, style, and stability.

·Features·FAQ

Documentation

Voice ID *

Voice ID to use for speech generation

Text Prompt *

Text to convert to speech

Sound ID

Background sound ID (optional)

Similarity Boost

Similarity boost value (0-100)

Style

Style value (0-100)

Speed

Speech speed multiplier (0.5-2.0)

Stability

Voice stability value (0-100)

Output

Your generated audio will appear here

Features

What Higgsfield TTS API offers

voice_id from the Higgsfield voices resource list

prompt string carries the full text to speak

Optional sound_id background bed from the Higgsfield sounds catalog

similarity_boost 0 to 100 to lock timbre to the base voice

style 0 to 100 for expressiveness emphasis

speed multiplier capped between 0 and 1.2 in the current form

stability 0 to 100 for run to run consistency

Uses Higgsfield TTS and feed polling endpoints documented on Unifically

Use cases

Built for

Primary

Narration with optional atmospheric bed from the sound_id picker

Product explainers when you need speed and style fine tuning

Prototype voice overs before committing to longer ElevenLabs sessions

Short social VO where bundled ambience saves a second pass in a DAW

Game barks with stable takes when you raise stability

Draft audiobook snippets with similarity tuning before master processing

FAQ

About Higgsfield TTS API

You POST voice_id and prompt. Optional sound_id adds background. Numeric controls map to similarity_boost, style, speed, and stability in the published schema.

Higgsfield TTS is listed at $0.05 per generation in Unifically pricing alongside other Higgsfield audio items.

Similarity_boost and style use 0 to 100. Stability uses 0 to 100. Speed is capped at 1.2 times in the slider with finer values accepted by the API handler.

The playground loads Higgsfield voices and sounds from the Unifically resource endpoints so labels stay in sync with what your key can access.