f5-tts
F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching
Decision gist · record as of 2026-08-14
Yes, if you have GPU hardware and need flexible, high-quality speech synthesis with voice cloning. The package is actively maintained, has no known vulnerabilities, and offers both interactive and programmatic interfaces. Install friction is low. However, it is not suitable for CPU-only environments due to inference speed, and requires careful PyTorch setup for your specific GPU architecture (NVIDIA, AMD, Intel, or Apple Silicon).AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires PyTorch with GPU support (NVIDIA, AMD, Intel, or Apple Silicon) for practical inference speed; CPU-only inference will be very slow.
- Minimum Python 3.10 recommended.
- Low install friction with a pure Python wheel.
License · maintenance · safety
MIT License (permissive) — MIT License permits unrestricted use, modification, and distribution with minimal restrictions—suitable for both research and commercial applications.
last release 2026-07-23 (22 days) · last repo commit 2026-07-23 · 15,119 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 104,785 downloads/mo, #12,737 on PyPI
Alternatives
Verify before relying
pip install f5-tts
from f5_tts.infer.api import F5TTS
model = F5TTS()
audio = model.infer(
ref_audio='prompt.wav',
ref_text='The content of reference audio.',
gen_text='Text to synthesize.'
)- Whether the package supports real-time or streaming inference beyond the documented batch/offline modes.
- Specific language support beyond the mentioned Chinese-English bilingual base model.
- Memory requirements for different model sizes and batch configurations.
- Whether fine-tuning is practical on consumer hardware or requires enterprise-grade GPUs.
What it is and what it does
F5-TTS is a neural text-to-speech system that converts written text into natural-sounding speech using diffusion transformers with flow matching. It accepts a reference audio clip and its transcription to learn a speaker's voice characteristics, then generates new speech in that voice for arbitrary text input. The package includes both a Diffusion Transformer variant (F5-TTS) optimized for speed and a Flat-UNet variant (E2-TTS) for closer paper reproduction.
The package ships with a Gradio web interface for interactive use, a command-line tool for batch inference, and a Python API for programmatic access. It supports multi-speaker synthesis, style transfer, and voice chat features. Inference is accelerated via PyTorch on GPU hardware; the fact sheet documents deployment via Triton and TensorRT-LLM for production use. Training and fine-tuning are supported through Hugging Face Accelerate.
Use it for
- Generate natural-sounding narration or audiobook content from text while preserving a specific speaker's voice characteristics.
- Clone a speaker's voice from a short reference clip and synthesize new dialogue in that voice for video dubbing or animation.
- Build a voice chat application that responds with synthesized speech matching a user's preferred speaker style.
- Fine-tune the base model on custom speech data to improve synthesis quality for specialized domains or accented speech.
- Deploy a production TTS service using the documented Triton runtime for low-latency, high-concurrency inference.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you have GPU hardware and need flexible, high-quality speech synthesis with voice cloning.
The package is actively maintained, has no known vulnerabilities, and offers both interactive and programmatic interfaces. Install friction is low. However, it is not suitable for CPU-only environments due to inference speed, and requires careful PyTorch setup for your specific GPU architecture (NVIDIA, AMD, Intel, or Apple Silicon).
Install
f5-tts on PyPI
Before you install
Low install friction with a pure Python wheel. Active maintenance with recent updates (22 days since last release). Requires 28 runtime dependencies including PyTorch, torchaudio, and transformers; GPU support is strongly recommended for practical inference speed.
Requires PyTorch with GPU support (NVIDIA, AMD, Intel, or Apple Silicon) for practical inference speed; CPU-only inference will be very slow. Minimum Python 3.10 recommended.
License in practice
MIT License permits unrestricted use, modification, and distribution with minimal restrictions—suitable for both research and commercial applications.
Quickstart
pip install f5-tts
from f5_tts.infer.api import F5TTS
model = F5TTS()
audio = model.infer(
ref_audio='prompt.wav',
ref_text='The content of reference audio.',
gen_text='Text to synthesize.'
)Verify before relying
- Whether the package supports real-time or streaming inference beyond the documented batch/offline modes.
- Specific language support beyond the mentioned Chinese-English bilingual base model.
- Memory requirements for different model sizes and batch configurations.
- Whether fine-tuning is practical on consumer hardware or requires enterprise-grade GPUs.
Package facts
| License | MIT License permissive |
| Python support | Not specified |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 28 packagesacceleratebitsandbytescached_pathclickdatasetsema_pytorchgradiohydra-corelibrosamatplotlibnumpypydubpypinyinrjiebasafetensorssoundfiletomlitorchtorchaudiotorchcodectorchdiffeqtqdmtransformerstransformers_stream_generatorunidecodevocoswandbx_transformers |
| Maintenance | Actively maintained 22 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 104,785 / month, #12,737 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | License :: OSI Approved :: MIT LicenseOperating System :: OS IndependentProgramming Language :: Python :: 3 |
Evidence: f5_tts-1.1.22-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “multi-speaker TTS”
- f5-ttsF5-TTS generates natural-sounding speech from text using…
- TTSTTS is a deep learning library for text-to-speech synthesis that…
- coqui-ttsCoqui TTS synthesizes speech from text using deep learning models,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also anima-python · omnivoice · vocos · diffusers · coqui-tts · TTS · nemo-text-processing · chatterbox-tts · tomesd · k-diffusion