F5-TTS at a glance
Checked 15 Oct 2026F5-TTS is a diffusion-transformer (flow matching) text-to-speech model that clones a voice from a reference clip, with a Gradio app, CLI, Docker image and a Triton/TensorRT-LLM deployment path. The code is MIT, but the pre-trained weights are CC-BY-NC-4.0 because of the Emilia training data, so the released models cannot be used commercially. PyTorch runs on NVIDIA, AMD (Linux), Intel GPUs and Apple Silicon. Minimum hardware is not documented.
RepoQuarry. “F5-TTS.” Checked 15 Oct 2026. https://repoquarry.dev/projects/f5-ttsHow much of ElevenLabs does it replace?
2 of 4 jobs. Speech generation and zero-shot voice cloning, but released weights are non-commercial (CC-BY-NC-4.0).
For speech recognition, voice agents, see other ElevenLabs alternatives. You may also need to build an interface, an API server and hosting yourself.
How much of Descript does it replace?
2 of 3 jobs. Voice cloning and text-to-speech only; no transcript-based audio or video editing.
For transcript-based editing, see other Descript alternatives. You may also need to build an interface, an API server and hosting yourself.
Facts, with sources
Every answer links to where we found it. We read the documentation; we did not install it.
The pre-trained models are licensed under the CC-BY-NC license due to the training data Emilia, which is an in-the-wild dataset.huggingface.co · F5-TTS ↗ 1d ago
The pre-trained models are licensed under the CC-BY-NC license due to the training data Emilia, which is an in-the-wild dataset.github.com · F5-TTS#license ↗ 1d ago
Decoding on a single L20 GPU, using 26 different prompt_audio & target_text pairs, 16 NFE.github.com · F5-TTS#runtime ↗ 1d ago
Install pytorch with your ROCm version (Linux only)github.com · F5-TTS#install-pytorch-with-matched-device ↗ 1d ago