Instructions to use OpenASR/redimnet2-b6-cn with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenASR
How to use OpenASR/redimnet2-b6-cn with OpenASR:
# Install the openasr CLI: https://github.com/QuintinShaw/openasr/releases openasr pull redimnet2-b6-cn openasr transcribe audio.wav --model redimnet2-b6-cn
- Notebooks
- Google Colab
- Kaggle
docs: refresh card for fp16-only ship
Browse files
README.md
CHANGED
|
@@ -51,12 +51,9 @@ openasr transcribe meeting.wav --model xasr-zh-en --diarize --format srt
|
|
| 51 |
| Quant | File (`.oasr`) | Size |
|
| 52 |
|:------|:---------------|-----:|
|
| 53 |
| fp16 | `redimnet2-b6-cn-fp16.oasr` | 28 MB |
|
| 54 |
-
| q8_0 | `redimnet2-b6-cn-q8_0.oasr` | 27 MB |
|
| 55 |
-
| f32 | `redimnet2-b6-cn-f32.oasr` | 50 MB |
|
| 56 |
|
| 57 |
-
<sub>Single
|
| 58 |
-
parity
|
| 59 |
-
quantization is produced.</sub>
|
| 60 |
|
| 61 |
## 🧠 About ReDimNet2-B6 Speaker Embedder (CN-enhanced)
|
| 62 |
|
|
@@ -79,8 +76,8 @@ openasr model-pack import redimnet2 <src>.safetensors <out>.oasr \
|
|
| 79 |
--package-id redimnet2-b6-cn
|
| 80 |
```
|
| 81 |
|
| 82 |
-
The `.oasr` container is GGUF-backed;
|
| 83 |
-
|
| 84 |
|
| 85 |
## ⚖️ License
|
| 86 |
|
|
|
|
| 51 |
| Quant | File (`.oasr`) | Size |
|
| 52 |
|:------|:---------------|-----:|
|
| 53 |
| fp16 | `redimnet2-b6-cn-fp16.oasr` | 28 MB |
|
|
|
|
|
|
|
| 54 |
|
| 55 |
+
<sub>Single **fp16** build: projection weights ship as fp16; norms/biases and other
|
| 56 |
+
parity-sensitive tensors stay f32 inside the pack. No extra public quant tiers.</sub>
|
|
|
|
| 57 |
|
| 58 |
## 🧠 About ReDimNet2-B6 Speaker Embedder (CN-enhanced)
|
| 59 |
|
|
|
|
| 76 |
--package-id redimnet2-b6-cn
|
| 77 |
```
|
| 78 |
|
| 79 |
+
The `.oasr` container is GGUF-backed; projection weights are stored as fp16 while
|
| 80 |
+
norms/biases and other parity-sensitive tensors remain f32.
|
| 81 |
|
| 82 |
## ⚖️ License
|
| 83 |
|