Update README.md
Browse files
README.md
CHANGED
@@ -83,15 +83,15 @@ Kotoba-whisper-v1.1 improves the punctuation and the timestamp of the output fro
|
|
83 |
we need to obtain the timestamps, which decreases the latency of the original kotoba-whisper-v1.0. See the following table comparing the inference speed on
|
84 |
transcribing **50min** Japanese speech audio, where we report the average over five independent runs.
|
85 |
|
86 |
-
| model
|
87 |
-
|
88 |
-
| kotoba-tech/kotoba-whisper-v1.0
|
89 |
-
| kotoba-tech/kotoba-whisper-v1.0
|
90 |
-
| kotoba-tech/kotoba-whisper-v1.1
|
91 |
-
| kotoba-tech/kotoba-whisper-v1.1 | True |
|
92 |
-
| kotoba-tech/kotoba-whisper-v1.1 | True |
|
93 |
-
| openai/whisper-large-v3
|
94 |
-
| openai/whisper-large-v3
|
95 |
|
96 |
|
97 |
## Transformers Usage
|
|
|
83 |
we need to obtain the timestamps, which decreases the latency of the original kotoba-whisper-v1.0. See the following table comparing the inference speed on
|
84 |
transcribing **50min** Japanese speech audio, where we report the average over five independent runs.
|
85 |
|
86 |
+
| model | return_timestamps | time (mean) |
|
87 |
+
|:---------------------------------------------------------|:--------------------|--------------:|
|
88 |
+
| kotoba-tech/kotoba-whisper-v1.0 | False | 10.8 |
|
89 |
+
| kotoba-tech/kotoba-whisper-v1.0 | True | 15.7 |
|
90 |
+
| kotoba-tech/kotoba-whisper-v1.1 (punctuator + stable-ts) | True | 17.9 |
|
91 |
+
| kotoba-tech/kotoba-whisper-v1.1 (punctuator) | True | 17.7 |
|
92 |
+
| kotoba-tech/kotoba-whisper-v1.1 (stable-ts) | True | 16.1 |
|
93 |
+
| openai/whisper-large-v3 | False | 29.1 |
|
94 |
+
| openai/whisper-large-v3 | True | 37.9 |
|
95 |
|
96 |
|
97 |
## Transformers Usage
|