| 1 | [build-system] |
| 2 | requires = ["setuptools >= 61.0", "setuptools-scm>=8.0"] |
| 3 | build-backend = "setuptools.build_meta" |
| 4 | |
| 5 | [project] |
| 6 | name = "f5-tts" |
| 7 | version = "1.1.22" |
| 8 | description = "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching" |
| 9 | readme = "README.md" |
| 10 | license = {text = "MIT License"} |
| 11 | classifiers = [ |
| 12 | "License :: OSI Approved :: MIT License", |
| 13 | "Operating System :: OS Independent", |
| 14 | "Programming Language :: Python :: 3", |
| 15 | ] |
| 16 | dependencies = [ |
| 17 | "accelerate>=0.33.0", |
| 18 | "bitsandbytes>0.37.0; platform_machine!='arm64' and platform_system!='Darwin'", |
| 19 | "cached_path", |
| 20 | "click", |
| 21 | "datasets", |
| 22 | "ema_pytorch>=0.5.2", |
| 23 | "gradio>=6.15.0", |
| 24 | "hydra-core>=1.3.0", |
| 25 | "librosa", |
| 26 | "matplotlib", |
| 27 | "numpy<=1.26.4; python_version<='3.10'", |
| 28 | "pydub", |
| 29 | "pypinyin", |
| 30 | "rjieba", |
| 31 | "safetensors", |
| 32 | "soundfile", |
| 33 | "tomli", |
| 34 | "torch>=2.0.0", |
| 35 | "torchaudio>=2.0.0", |
| 36 | "torchcodec", |
| 37 | "torchdiffeq", |
| 38 | "tqdm>=4.65.0", |
| 39 | "transformers", |
| 40 | "transformers_stream_generator", |
| 41 | "unidecode", |
| 42 | "vocos", |
| 43 | "wandb", |
| 44 | "x_transformers>=1.31.14", |
| 45 | ] |
| 46 | |
| 47 | [project.optional-dependencies] |
| 48 | eval = [ |
| 49 | "faster_whisper==0.10.1", |
| 50 | "funasr", |
| 51 | "jiwer", |
| 52 | "modelscope", |
| 53 | "zhconv", |
| 54 | "zhon", |
| 55 | ] |
| 56 | |
| 57 | [project.urls] |
| 58 | Homepage = "https://github.com/SWivid/F5-TTS" |
| 59 | |
| 60 | [project.scripts] |
| 61 | "f5-tts_infer-cli" = "f5_tts.infer.infer_cli:main" |
| 62 | "f5-tts_infer-gradio" = "f5_tts.infer.infer_gradio:main" |
| 63 | "f5-tts_finetune-cli" = "f5_tts.train.finetune_cli:main" |
| 64 | "f5-tts_finetune-gradio" = "f5_tts.train.finetune_gradio:main" |
| 65 |