whisperkittools generated README.md
Browse files
README.md
CHANGED
@@ -15,8 +15,6 @@ tags:
|
|
15 |
|
16 |
## Dataset: `librispeech`
|
17 |
|
18 |
-
### Quality Evaluation
|
19 |
-
|
20 |
| | WER | QoI (%) | File Size (MB) |
|
21 |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------:|----------:|-----------------:|
|
22 |
| [WhisperOpenAIAPI/openai_whisper-large-v2](https://hf.co/datasets/argmaxinc/whisperkit-evals/tree/main/WhisperOpenAIAPI/openai_whisper-large-v2/librispeech) | 2.85 | 100 | 3100 |
|
@@ -35,7 +33,7 @@ tags:
|
|
35 |
| [WhisperKit/openai_whisper-tiny](https://hf.co/datasets/argmaxinc/whisperkit-evals/tree/main/WhisperKit/openai_whisper-tiny/librispeech) | 8.94 | 52.4 | 66 |
|
36 |
|
37 |
|
38 |
-
We believe that rigorously measuring the
|
39 |
enterprises to make informed decisions when opting to use optimized or compressed variants of
|
40 |
any machine learning model in production. To contextualize `WhisperKit`, we take the following Whisper
|
41 |
implementations and benchmark them using a consistent evaluation harness:
|
|
|
15 |
|
16 |
## Dataset: `librispeech`
|
17 |
|
|
|
|
|
18 |
| | WER | QoI (%) | File Size (MB) |
|
19 |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------:|----------:|-----------------:|
|
20 |
| [WhisperOpenAIAPI/openai_whisper-large-v2](https://hf.co/datasets/argmaxinc/whisperkit-evals/tree/main/WhisperOpenAIAPI/openai_whisper-large-v2/librispeech) | 2.85 | 100 | 3100 |
|
|
|
33 |
| [WhisperKit/openai_whisper-tiny](https://hf.co/datasets/argmaxinc/whisperkit-evals/tree/main/WhisperKit/openai_whisper-tiny/librispeech) | 8.94 | 52.4 | 66 |
|
34 |
|
35 |
|
36 |
+
We believe that rigorously measuring the quality of inference is necessary for developers and
|
37 |
enterprises to make informed decisions when opting to use optimized or compressed variants of
|
38 |
any machine learning model in production. To contextualize `WhisperKit`, we take the following Whisper
|
39 |
implementations and benchmark them using a consistent evaluation harness:
|