- Bump gradio version from 5.0 to 6.0, with several fixes to ensure compatibility with f5-tts_infer-gradio **Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.10...1.1.12

- Fix issues of skipping last words in shorter sentences when batch inference #1039 #1179, in both runtime TRT-LLM and PyTorch implementation. - Clean-up and make advanced eval scripts 529d8561331a6d0a3dcfa3ff7da60679aeb86ac1 - Add torchcodec dependency to pyproject.toml - Replace jieba package with rjieba #1230 **Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.9...1.1.10

- Update numpy version constraint to support numpy 2.x for python version >= 3.11 #1162 **Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.8...1.1.9

**Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.7...1.1.8

## What's Changed * Add flash attn2 support attn_mask, fix some minor bug in eval by @ZhikangNiu in https://github.com/SWivid/F5-TTS/pull/1066 * Speedup inference by batching CFG in DiT by @starkwj in https://github.com/SWivid/F5-TTS/pull/1084 * Update WAV File Naming and Dependencies 📝🔊 by @danhtran2mind in https://github.com/SWivid/F5-TTS/pull/1091 ## New Contributors * @starkwj made their first contribution in https://github.com/SWivid/F5-TTS/pull/1084 * @danhtran2mind made their first contribution in https://github.com/SWivid/F5-TTS/pull/1091 **Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.5...1.1.6

## What's Changed * German Model support by @hvoss-techfak in https://github.com/SWivid/F5-TTS/pull/1072 * Fix finetune training with spaces in file paths by @ionite34 in https://github.com/SWivid/F5-TTS/pull/1063 * Fix the duration computation in `triton_trtllm/client_grpc.py` by @Nietism in https://github.com/SWivid/F5-TTS/pull/1071 * Update: Empirically Pruned Step Sampling by @Jerrister in https://github.com/SWivid/F5-TTS/pull/1077 ## New Contributors * @hvoss-techfak made their first contribution in https://github.com/SWivid/F5-TTS/pull/1072 * @ionite34 made their first contribution in https://github.com/SWivid/F5-TTS/pull/1063 * @Nietism made their first contribution in https://github.com/SWivid/F5-TTS/pull/1071 **Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.4...1.1.5

**Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.3...1.1.4 - Bug Fix for #1015 dad398c0c164abdcbe6cc111292aa83f127274fc - Minor Fix for backward compatibility 3d969bf78dcabfc75acd4a0ecdcd9f0fd856ee8b

**Full Changelog**: https://github.com/SWivid/F5-TTS/compare/1.1.0...1.1.3 - v1.1.3 better gradio inference ui, with cherrypick, cache and load text from file support - v1.1.2 update data prepare script; fix speech edit; fix tensorrt-llm server code-switch inference - v1.1.1 add benchmark and trtllm offline code Thanks to @yuekaizhang @fakerybakery @petermg @TBXark