Georgi Gerganov
Georgi Gerganov
Try to make a very basic watchOS example app and see if we can run at least the tiny model efficiently. The [iOS example](https://github.com/ggerganov/whisper.cpp/discussions/97) could be a good starting point....
This is an attempt to simplify and unify the SIMD implementation for various architectures. The idea is similar to https://github.com/ggerganov/whisper.cpp/pull/95#issuecomment-1344845687 We define a common subset of C macros which map...
ref #200 Instead of creating and joining new threads for each graph compute, we create a thread pool with active threads and make them wait on a condition variable instead...
The idea is to avoid keeping all intermediate tensors of the computation graph by introducing "scratch" buffers to `ggml` https://github.com/ggerganov/whisper.cpp/issues/272#issuecomment-1353739709 I initially thought it would be enough to just keep...
Just had an idea for a new example app that allows you to make Chess moves using voice. The example will run in a web-page and will be based on...
See notebook, section "Word-level timestamps using attention weights": https://github.com/openai/whisper/blob/main/notebooks/Multilingual_ASR.ipynb
close #37 @mikeslattery Could you please check if this works on MUSL Linux?
Last time I checked, the results produced by `whisper.cpp` for computing the Log-Mel spectrogram were not exactly identical to the OpenAI implementation: - `whisper.cpp`: https://github.com/ggerganov/whisper.cpp/blob/master/whisper.cpp#L2284-L2298 - OpenAI Whisper: https://github.com/openai/whisper/blob/main/whisper/audio.py#L92-L124 I...
Makes sense - hopefully will reduce hallucinations https://github.com/openai/whisper/discussions/838#discussioncomment-5222689
Running Whisper inference on Apple Neural Engine (ANE) via Core ML Huge thanks to @wangchou for demonstrating how to use Core ML and making the initial port: https://github.com/ggerganov/whisper.cpp/discussions/548 **WIP IN...