Add sherpa-onnx int8 runtime files for offline desktop inference
Browse filesAdditive only: large-v3 encoder/decoder int8 ONNX + tokens. Existing weights untouched. Source: csukuangfj/sherpa-onnx-whisper-large-v3. Heavy model, 8/16GB RAM.
- large-v3-decoder.int8.onnx +3 -0
- large-v3-encoder.int8.onnx +3 -0
- large-v3-tokens.txt +0 -0
large-v3-decoder.int8.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebc6bfd88e162a46cb3edee8a7e727e1dcbc65cabecb19e2573695e4d495e1af
|
| 3 |
+
size 1008265203
|
large-v3-encoder.int8.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d531cf17248acc43e8c09b472a0877055e770877857a5332fc1304b36534ec85
|
| 3 |
+
size 766671985
|
large-v3-tokens.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|