Skip to content

discussions Search Results · repo:ggerganov/whisper.cpp language:CMake

Filter by

359 results
 (46 ms)

359 results

inggerganov/whisper.cpp (press backspace or delete to remove)

ggerganov/whisper.cpp
: 
Word level time stamps with Whisper 1.6.2 / DTW off - solutions?

Hello Firstly WhisperCPP has been awesome to work with. Im using the Swift API and am experimenting with the word level time stamps. Im noticing that the timing is quite off, and im curious how the params ...
  • vadeposted
    yesterday
  • 0

ggerganov/whisper.cpp
: 
Are the CoreML models downloadable?

The script download-coreml-model.sh is no longer functional, but can I download them manually somehow? I really struggling to convert them myself. I keep getting the error: Traceback (most recent call ...

ggerganov/whisper.cpp
: 
Faster than distilled whisper? Could run distilled models?

Hello, was wondering if this is faster than the distilled models from https://huggingface.co/distil-whisper ? Or could it even potentionally use those distilled models for even faster processing ? Thanks! ...

ggerganov/whisper.cpp
: 
Real-time transcription on Raspberry Pi 4

It is possible to some extend to run Whisper in real-time mode on an embedded device such as the Raspberry Pi. Below are a few examples + build instructions. Real-time with 4 seconds step ./stream -m ...

ggerganov/whisper.cpp
: 
How to increase the likelihood of catching voice command?

I have tried of the command list method and grammar method in the examples. both did not achieve desired results. for the command list method it seems only work for single word command. And for the grammar ...

ggerganov/whisper.cpp
: 
Diarization in Rust

What do you think about getting word segments by settings max_len to 1 with split_on_word and then go through the words and run voice embedding model on that segment to get speaker label? I m thinking ...

ggerganov/whisper.cpp
: 
How to reduce the vocabulary size of ggml model?

I am working on a speech to command program where only a handful of commands is needed. So how do I reduce the vocabulary size so the it is faster to process and also have higher change of catching the ...

ggerganov/whisper.cpp
: 
Question about the `embed_positions.weight` in Whisper model

In the encoder of Whisper models, the encoder.embed_positions.weight was initialized using sine and cosine functions of different frequencies. However, in the decoder of Whisper models, the decoder.embed_positions.weight ...

ggerganov/whisper.cpp
: 
Improving hallucinations and repetitions

I created this to write down some of my observations for improving output quality of whisper - especially regarding hallucinations and repetitions. This is from an attempt of translating an entire foreign ...

ggerganov/whisper.cpp
: 
'command' not using CUDA?

I cloned the whisper.cpp repository a few days ago. Initially, I did a plain make clean; make -j which generated a few warnings but otherwise seem to finish fine. I also downloaded all the models listed ...