You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,9 +4,9 @@ A proper C# wrapper for the **Fish Audio S2 Pro** repo [s2.cpp](https://github.c
4
4
What features are currently implemented:
5
5
- Ability to transform text to voice using either ***only using a text input*** and/or combined ***with an reference audio(voice sample) file***.
6
6
- Ability to save the resulting audio to ***file*** and/or retreive it ***as an array of float samples*** in code(mono).
7
-
- The library is modular, meaning you can load the model/tokenizer separately, which allows you to re-use the model/tokenizer in other FishS2Client instances without reloading them again.
8
-
- Audio references are binded and stored in FishS2Client instances, and they are processed at the time you create/add/register them with RegisterVoiceReference(). This allows to process the reference voice samples for the model ***ONLY ONCE*** and re-use it without having to re-load it for every prompt.
9
-
- CUDA, Vulkan and Metal support, as well as CPU fallback. (As long you have **ggml-cuda.dll**/**ggml-vulkan.dll**/**ggml-metal.dll** in the same folder as this library and the s2.dll)
7
+
- The library is modular, meaning you can load the model/clients separately, which allows you to re-use the model in other FishS2Client instances without reloading them again.
8
+
- Audio references are binded and stored in FishAudioVoiceReference instances, and they are processed at the time you create/add/register them with RegisterVoiceReference(). This allows to process the reference voice samples for the model ***ONLY ONCE*** and re-use it without having to re-load it for every prompt. (NOTE: You can only use/synth. processed voice samples for the model you used to process them with.)
9
+
- CUDA, Vulkan and Metal support, as well as CPU fallback. (As long you have **ggml-cuda.dll**/**ggml-vulkan.dll**/**ggml-metal.dll** in the same folder as this library and the **s2.dll**)
10
10
- This library is compiled against the netstandard 2.1, that means you can use it both in .NET 5+ applications and the Unity game engine, allowing this to be used for games as well.
11
11
12
12
You still have to build [s2.dll](https://github.com/rodrigomatta/s2.cpp) manually along with the ggml.
0 commit comments