SWivid / F5-TTS

Official code for "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"
https://arxiv.org/abs/2410.06885
MIT License
7.21k stars 859 forks source link

The value in this place does not match #461

Open hjj-lmx opened 5 days ago

hjj-lmx commented 5 days ago

Checks

Question details

api.py

1731503398765 1731503471480

SWivid commented 5 days ago
  1. As a pip package (if just for inference)
  2. Local editable (if also do training, finetuning)
hjj-lmx commented 5 days ago
  1. As a pip package (if just for inference)
  2. Local editable (if also do training, finetuning)

How to integrate it into your own project, which logic should be used as the standard, and whether it is possible to use the API. py file

SWivid commented 5 days ago
  1. vocab.txt is the same.
  2. pip install pkg way will only get the files inside src/, while local editable could take outside folder for own project usage
  3. make any modification up to your need
hjj-lmx commented 5 days ago
  1. vocab.txt is the same.
  2. pip install pkg way will only get the files inside src/, while local editable could take outside folder for own project usage
  3. make any modification up to your need

Is there a complete usage document available

SWivid commented 5 days ago

for what usage e.g.

hjj-lmx commented 5 days ago

用于什么用途例如

API like invocation; Now it's all done through commands

SWivid commented 5 days ago

https://github.com/SWivid/F5-TTS/tree/main/src/f5_tts/infer#socket-realtime-client see if it would help.