ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2024-12-18 20:27:53 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Georgi Gerganov 2022-12-07 05:15:46 +02:00 committed by GitHub
parent faa85f9840
commit 3996ecc156
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 15 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
README.md
View File

@ -52,21 +52,6 @@ The tensor operators are optimized heavily for Apple silicon CPUs. Depending on
instrisics or CBLAS Accelerate framework routines are used. The latter are especially effective for bigger sizes since instrisics or CBLAS Accelerate framework routines are used. The latter are especially effective for bigger sizes since
the Accelerate framework utilizes the special-purpose AMX coprocessor available in modern Apple products. the Accelerate framework utilizes the special-purpose AMX coprocessor available in modern Apple products.
## Limitations
- Inference only
- No GPU support
- Very basic greedy sampling scheme - always pick up the token with highest probability.
This should be similar to the [GreedyDecoder](https://github.com/openai/whisper/blob/main/whisper/decoding.py#L249-L274)
from the original python implementation, so in order to make a fair comparison between the 2 implementations, make sure
to run the python code with the following parameters:
```
whisper --best_of None --beam_size None ...
```
In the future, `whisper.cpp` will support more sampling strategies.
## Quick start ## Quick start
First, download one of the Whisper models converted in [ggml format](models). For example: First, download one of the Whisper models converted in [ggml format](models). For example:
@ -220,6 +205,21 @@ make large
| medium | 1.5 GB | ~2.6 GB | `fd9727b6e1217c2f614f9b698455c4ffd82463b4` | | medium | 1.5 GB | ~2.6 GB | `fd9727b6e1217c2f614f9b698455c4ffd82463b4` |
| large | 2.9 GB | ~4.7 GB | `0f4c8e34f21cf1a914c59d8b3ce882345ad349d6` | | large | 2.9 GB | ~4.7 GB | `0f4c8e34f21cf1a914c59d8b3ce882345ad349d6` |
## Limitations
- Inference only
- No GPU support
- Very basic greedy sampling scheme - always pick up the token with highest probability.
This should be similar to the [GreedyDecoder](https://github.com/openai/whisper/blob/main/whisper/decoding.py#L249-L274)
from the original python implementation, so in order to make a fair comparison between the 2 implementations, make sure
to run the python code with the following parameters:
```
whisper --best_of None --beam_size None ...
```
In the future, `whisper.cpp` will support more sampling strategies.
## Another example ## Another example
Here is another example of transcribing a [3:24 min speech](https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg) Here is another example of transcribing a [3:24 min speech](https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg)