FreeTurn English speech into punctuated text with word timestamps using an NVIDIA open ASR model.
Audio & Voice
All tools
View more Free 
Free tier
Free
Access
Listed as Fugatto is an NVIDIA research model and the official research page links to a public demo. The current source does not present a stable consumer plan or production API, so the demo should be treated as free research access where available rather than unlimited hosted generation.
0
Audio & Voice Checked Aug 6, 2026
Fugatto
Use text and optional audio to explore unusual sound generation and transformation.
Best for: Musicians, sound designers, researchers, and curious creators exploring controllable generative audio.
What you can do with it
- Generate a sound from a written description
- Transform an audio clip with a text instruction
- Explore unusual sound combinations
- Prototype a sound-design direction
- Study text-audio compositional control
Good to know
- Text and audio conditioning
- Unusual sound transformation research
- Official NVIDIA demo link
- Strong creative-audio use case
Limitations
- Research access can change
- No stable consumer plan shown
- Output needs rights and quality review
Other official links
Keep exploring
Similar tools
Free Free tierTransform a live sound into a new playable texture while keeping its musical shape.
Audio & Voice
FreeRun multilingual speech recognition or translation with an open NVIDIA model.
Audio & Voice