VTS

Spec sheet

Purpose

VTS generates sound effects from a short vocal or audio sketch combined with a text prompt.Read 4 Oct 2026 · github.com

Intended users

The checkpoint is intended for research and creative sound-effect generation from vocal sketches or short audio sketches plus text prompts.Read 4 Oct 2026 · huggingface.co

Deployment

The repository provides an inference-only package with local model and runtime code; training code and datasets are excluded.Read 4 Oct 2026 · github.com

Integrations

The inference code uses google/flan-t5-base as its text encoder and includes local vocoder code.Read 4 Oct 2026 · github.com

Support

The repository says to contact the maker at [email protected] with questions.Read 4 Oct 2026 · github.com

License

The project and model checkpoint are listed under the MIT License.Read 4 Oct 2026 · github.com

API availability

The model is not packaged as a Hugging Face Inference API pipeline and is not deployed by an Inference Provider.Read 4 Oct 2026 · huggingface.co

Checkpoint

The inference code uses the pretrained checkpoint named dynamic_v3_0415.ckpt, which is available from the linked Hugging Face model repository.Read 4 Oct 2026 · github.com

Checkpoint access

If the Hugging Face repository requires authentication, inference setup uses an HF_TOKEN environment variable.Read 4 Oct 2026 · github.com

Generation

Generated WAV files are written to a chosen output directory, and the default output duration follows the input audio duration unless a duration is specified.Read 4 Oct 2026 · github.com

Generation length

Output duration defaults to the input audio duration, can be configured, and the checkpoint is tuned for short sound-effect clips.Read 4 Oct 2026 · github.com

Hardware

The quick-start inference example specifies the CUDA device.Read 4 Oct 2026 · github.com

Hardware setup

The documented local requirements pin PyTorch and torchaudio CUDA 12.4 builds, with instructions to install matching builds for other CUDA drivers.Read 4 Oct 2026 · github.com

How it works

It uses voice conditioning derived from dynamic audio features alongside text conditioning from a prompt.Read 4 Oct 2026 · github.com

Limits

The model is optimized for short sound-effect style clips, and output quality depends on the checkpoint, input audio, prompt text, and sampling settings.Read 4 Oct 2026 · huggingface.co

Local inference

The repository provides an inference-only package with local model and runtime code.Read 4 Oct 2026 · github.com

Output

Generated audio files are written as WAV files to the chosen output directory.Read 4 Oct 2026 · github.com

Sampling

Sampling uses a local ODE solver and typically runs 64 steps with CFG scale 3.0.Read 4 Oct 2026 · github.com

Text encoder

The inference path encodes text prompts with google/flan-t5-base.Read 4 Oct 2026 · github.com

Training

Training code and dataset manifests are not included in the inference package.Read 4 Oct 2026 · github.com

Voice conditioning

Voice conditioning uses dynamic features derived from spectral centroid, RMS, and chroma-index signals.Read 4 Oct 2026 · github.com

Where Specifiction places it

• AI Sound Effect Generators#29 of 296.6