heimoshuiyu/faster-whisper
1
0
Fork 0
Code Issues Pull Requests 1 Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Guillaume Klein
2023-02-13 17:43:22 +01:00
committed by GitHub
parent 90f6923be0
commit 47a62ab975
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
README.md
View File

@@ -57,12 +57,12 @@ for segment in segments:
print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))
```
## Comparing performance against openai/whisper
## Comparing performance against other implementations
If you are comparing the performance against [openai/whisper](https://github.com/openai/whisper), you should make sure to use the same settings in both frameworks. In particular:
If you are comparing the performance against other Whisper implementations, you should make sure to run the comparison with similar settings. In particular:
* In openai/whisper, `model.transcribe` uses a beam size of 1 by default. A different beam size will have an important impact on performance so make sure to use the same.
* When running on CPU, make sure to set the same number of threads. Both frameworks will read the environment variable `OMP_NUM_THREADS`, which can be set when running your script:
* Verify that the same transcription options are used, especially the same beam size. For example in openai/whisper, `model.transcribe` uses a default beam size of 1 but here we use a default beam size of 5.
* When running on CPU, make sure to set the same number of threads. Many frameworks will read the environment variable `OMP_NUM_THREADS`, which can be set when running your script:
```bash
OMP_NUM_THREADS=4 python3 my_script.py