AUDIO
Give your application a voice
Create expressive speech and complete sound scenes through LSF’s Audio API. Direct the voice, emotion, timing, ambience and sound effects from one prompt, with supported audio or image references when needed.
SEED AUDIO 1.0
Direct the whole sound scene
Shape speech, performance, ambience and sound cues as one coordinated result. Guide emotion, pacing, timing and background texture from a single creative direction.
SEED AUDIO 1.0 IN ACTION
Hear control, expression and timing
The same voice can move across very different performance styles, while prompt-level timing controls help each line enter at the intended moment.
01 · EXPRESSIVE VOICE
One voice, varied styles
Keep a recognizable voice identity while changing emotion, rhythm and delivery for commentary, audiobooks, live commerce and other creative formats.
02 · TIMING CONTROL
Place dialogue at the right moment
Specify when character lines should enter, with current dialogue timing control at 100 ms intervals for dubbing, advertising and scripted audio.
ONE CREATIVE SYSTEM
From a line of dialogue to a complete audio scene
Create a voice from a text description, an authorized reference sample, or both—then keep its identity consistent while the performance changes.
Performance control
Direct emotion, prosody, pacing and speaking style without losing the intended voice identity.
Scene-level sound
Coordinate dialogue with ambience, effects and background texture instead of assembling isolated tracks.
Multilingual expression
Build experiences across supported languages while keeping the creative direction coherent.
Longer continuity
Generate up to two minutes in one pass and continue the result for longer-form audio workflows.
WORKFLOW
Describe, reference and generate
Write the direction
Define the words, language, delivery and sound environment you want.
Add approved context
Include supported image or authorized audio references when the request mode allows them.
Save the result
Generate the audio, read the returned URL and move the file to your own storage before it expires.
Text to speech
Supply the text to read and supported delivery instructions.
Reference-aware requests
Use public reference URLs with the supported request mode. Check the reference and combination requirements in the API guide.
Clear delivery
Read the returned audio URL and save the file to your own storage before the link expires.
Audio has its own pricing
Seed Audio 1.0 is $0.18 per generated minute, billed by output duration with one-second precision. LSF list price; tiered pricing available. Contracted customer pricing may differ.
View Audio pricingBUILD WITH LSF
Start your next video project with $20
Get $20 in free test credit. Tell us what you are building, and we’ll help configure your API access and first test.