speechbrain
/

asr-streaming-conformer-gigaspeech

Automatic Speech Recognition

Model card Files Files and versions Community

sdelangen commited on 16 days ago

Commit

d7a6b79

•

1 Parent(s): c7b4841

Update README.md

Files changed (1) hide show

README.md +14 -12

README.md CHANGED Viewed

@@ -48,18 +48,20 @@ The performance of the model in full context mode (no streaming) is the followin
 | Release | Test WER | GPUs |
 |:-------------:|:--------------:|:--------:|
-| 24-02-26 | TBD | 4xA100 40GB |
-With streaming, the results with different chunk sizes on test-clean are the following:
-|       | full | cs=32 (1280ms) | 24 (960ms) | 16 (640ms) | 12 (480ms) | 8 (320ms) |
-|:-----:|:----:|:-----:|:-----:|:-----:|:-----:|:-----:|
-| full  | TBD  | -     | -     | -     | -     | -     |
-| lc=32 | -    | TBD   | TBD   | TBD   | TBD   | TBD   |
-| 16    | -    | TBD   | TBD   | TBD   | TBD   | TBD   |
-| 8     | -    | TBD   | 11.50% | TBD   | TBD   | TBD   |
-| 4     | -    | TBD   | TBD   | TBD   | TBD   | TBD   |
-| 2     | -    | TBD   | TBD   | TBD   | TBD   | TBD   |
 ## Pipeline description

 | Release | Test WER | GPUs |
 |:-------------:|:--------------:|:--------:|
+| 24-02-26 | 11.00% | 4xA100 40GB |
+TODO: validate the results
+With streaming, the results with different chunk sizes on the test split are the following:
+|       | full   | cs=32 (1280ms) | 24 (960ms) | 16 (640ms) | 12 (480ms) | 8 (320ms) |
+|:-----:|:------:|:------:|:------:|:------:|:------:|:------:|
+| full  | 11.00% | -      | -      | -      | -      | -      |
+| lc=32 | -      | -      | TBD    | TBD    | TBD    | TBD    |
+| 16    | -      | -      | TBD    | TBD    | TBD    | TBD    |
+| 8     | -      | -      | 11.50% | TBD    | TBD    | TBD    |
+| 4     | -      | 11.40% | TBD    | TBD    | TBD    | TBD    |
+| 2     | -      | 11.46% | TBD    | TBD    | TBD    | TBD    |
 ## Pipeline description