Update README.md
Browse files
README.md
CHANGED
|
@@ -105,7 +105,7 @@ wespeaker - Speaker embedding techniques
|
|
| 105 |
- **Output**: Speaker segments with timestamps and IDs
|
| 106 |
- **Framework**: CoreML (converted from PyTorch)
|
| 107 |
- **Optimization**: Apple Neural Engine (ANE) optimized operations
|
| 108 |
-
- **Precision**: FP32
|
| 109 |
|
| 110 |
## Training Data
|
| 111 |
|
|
|
|
| 105 |
- **Output**: Speaker segments with timestamps and IDs
|
| 106 |
- **Framework**: CoreML (converted from PyTorch)
|
| 107 |
- **Optimization**: Apple Neural Engine (ANE) optimized operations
|
| 108 |
+
- **Precision**: FP32 on CPU/GPU, FP16 on ANE
|
| 109 |
|
| 110 |
## Training Data
|
| 111 |
|