mirror of
https://github.com/PABannier/bark.cpp
synced 2026-03-03 05:30:35 +01:00
DOC README file (#9)
This commit is contained in:
parent
4ab0f5a789
commit
8baeb1abdc
21
README.md
21
README.md
@ -1,6 +1,19 @@
|
||||
- [ ] CMakeLists
|
||||
- [ ] GitHub reference to GGML
|
||||
- [ ] Generate one binary large file for the weights
|
||||
# bark.cpp (coming soon!)
|
||||
|
||||
Inference of SunoAI's bark model in pure C/C++ using [ggml](https://github.com/ggerganov/ggml).
|
||||
|
||||
## Description
|
||||
|
||||
The main goal of `bark.cpp` is to synthesize audio from a textual input with the [Bark](https://github.com/suno-ai/bark) model using 4-bit quantization on a MacBook.
|
||||
|
||||
Bark has essentially 4 components:
|
||||
- [x] Semantic model to encode the text input
|
||||
- [x] Coarse model
|
||||
- [x] Fine model
|
||||
- [ ] Encoder (quantizer + decoder) to generate the waveform from the tokens
|
||||
|
||||
## Roadmap
|
||||
|
||||
- [ ] Quantization
|
||||
- [ ] FP16
|
||||
- [ ] Swift package
|
||||
- [ ] Swift package for iOS devices
|
||||
|
||||
Loading…
Reference in New Issue
Block a user