Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

sync : whisper.cpp #823

Merged
merged 14 commits into from
May 14, 2024
Merged

sync : whisper.cpp #823

merged 14 commits into from
May 14, 2024

Commits on May 14, 2024

  1. whisper : update grammar-parser.cpp (whisper/2058) 

    preceeding -> preceding
    @eltociear @ggerganov
    eltociear authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    8821421 View commit details
    Browse the repository at this point in the history

  2. main : pass nullptr when regex is empty (whisper/2070)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    6d3b994 View commit details
    Browse the repository at this point in the history

  3. whisper : more prominent log message for sub-1s audio (whisper/2065)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    85a1d20 View commit details
    Browse the repository at this point in the history

  4. Remove unnecessary memory reallocation in fft (whisper/2080) 

    fft_out needs to be twice the frame_size, not the frame_step.  It is resized in fft() anyway, but this change prevents an unnecessary reallocation.

n_fft must match the mel filter size, so it is best not to calculate it from the framesize.

We only need to get the magnitudes for half the spectrum since the other half is a mirror and not used in the mel filter loop later.
    @goldwaving @ggerganov
    goldwaving authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    623b291 View commit details
    Browse the repository at this point in the history

  5. ggml : expose SSE3 and SSSE3 for MSVC when AVX is available (whisper/… 

    …2128)
    @przemoc @ggerganov
    przemoc authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    083d952 View commit details
    Browse the repository at this point in the history

  6. ggml : try fix ppc64 (whisper/0)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    9466438 View commit details
    Browse the repository at this point in the history

  7. whisper : remove old flash attn code (whisper/0)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    14001e4 View commit details
    Browse the repository at this point in the history

  8. metal : tune soft_max number of threads (whisper/0)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    b842efa View commit details
    Browse the repository at this point in the history

  9. main : fix double quote escaping in csv output (whisper/2090)

    @mashizora @ggerganov
    mashizora authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    737407f View commit details
    Browse the repository at this point in the history

  10. whisper : switch back to F32 mask (whisper/0)

    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    71ce179 View commit details
    Browse the repository at this point in the history

  11. main : add options for temperature control (whisper/2088) 

    Add two options:

```
-tp,       --temperature N     [0.00   ] The sampling temperature, between 0 and 1
-tpi,      --temperature-inc N [0.20   ] The increment of temperature, between 0 and 1
```

The sampling temperature, between 0 and 1. Higher values like 0.8 will
make the output more random, while lower values like 0.2 will make it
more focused and deterministic. If set to 0, the model will use log
probability to automatically increase the temperature until certain
thresholds are hit.

Signed-off-by: Daniel Ziegenberg <[email protected]>
    @ziegenberg @ggerganov
    ziegenberg authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    36caae3 View commit details
    Browse the repository at this point in the history

  12. main : dont print timings with --no-prints (whisper/2108) 

    Signed-off-by: Daniel Ziegenberg <[email protected]>
    @ziegenberg @ggerganov
    ziegenberg authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    42526d6 View commit details
    Browse the repository at this point in the history

  13. whisper : fix model path encoding in windows (whisper/2086) 

    * fix: model path encoding in windows

* fix: convert model path to wide string only for MSVC compiler
    @thewh1teagle @ggerganov
    thewh1teagle authored and ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    c7385b4 View commit details
    Browse the repository at this point in the history

  14. sync : whisper.cpp 

    ggml-ci
    @ggerganov
    ggerganov committed May 14, 2024
    Configuration menu
    Copy the full SHA
    b2773a4 View commit details
    Browse the repository at this point in the history