card: FluidAudio #980 merged
Browse files
README.md
CHANGED
|
@@ -21,7 +21,7 @@ Every encoder here holds the checkpoint's **exact** five-value weights — as fp
|
|
| 21 |
`constexpr_sparse_to_dense`). No re-quantization noise is added, and all five produce identical transcripts. The decoder
|
| 22 |
and joint are re-exported from the checkpoint's int6 tables; preprocessor and vocabulary are the v3 ones.
|
| 23 |
|
| 24 |
-
## Usage (FluidAudio
|
| 25 |
|
| 26 |
```swift
|
| 27 |
let models = try await AsrModels.downloadAndLoad(version: .phonon2) // iOS 18+ / macOS 15+, downloads Encoder.mlmodelc
|
|
|
|
| 21 |
`constexpr_sparse_to_dense`). No re-quantization noise is added, and all five produce identical transcripts. The decoder
|
| 22 |
and joint are re-exported from the checkpoint's int6 tables; preprocessor and vocabulary are the v3 ones.
|
| 23 |
|
| 24 |
+
## Usage (FluidAudio, merged in [#980](https://github.com/FluidInference/FluidAudio/pull/980))
|
| 25 |
|
| 26 |
```swift
|
| 27 |
let models = try await AsrModels.downloadAndLoad(version: .phonon2) // iOS 18+ / macOS 15+, downloads Encoder.mlmodelc
|