whisper.cpp icon indicating copy to clipboard operation
whisper.cpp copied to clipboard

Port of OpenAI's Whisper model in C/C++

Results 525 whisper.cpp issues
Sort by recently updated
recently updated
newest added

Consider this command, likely wrong in many ways: ~/whisper.cpp/main -m ~/whisper.cpp/models/ggml-base.en.bin -f track1.wav -ml 1 -nt false --output-txt trk1.txt whisper_model_load: n_vocab = 51864 whisper_model_load: n_audio_ctx = 1500 whisper_model_load: n_audio_state =...

question

We now have a sample script for transcribing radio livestream, such as BBC radio: ```bash $ ./examples/livestream.sh http://a.files.bbci.co.uk/media/live/manifesto/audio/simulcast/hls/nonuk/sbr_low/ak/bbc_world_service.m3u8 10 small.en [+] Transcribing stream with model 'small.en', step_s 10 (press Ctrl+C...

good first issue
ideas

The output on the`whisper.wasm` demo for me says, ```cpp system_info: n_threads = 2 / 2 | AVX = 0 | AVX2 = 0 | AVX512 = 0 | NEON =...

bug

I got interested in this project because of its pure cpp and c implication. Mainly because of the idea of being able to port anywhere. So the question is, can...

question

CUDA toolkit documentation [link](https://docs.nvidia.com/cuda/nvblas/index.html) states that NVBLAS is a drop-in BLAS replacement. Also states: "The NVBLAS Library is a GPU-accelerated Library that implements BLAS (Basic Linear Algebra Subprograms). It can...

performance

whisper output [00:00.000 --> 00:04.000] 贵州省依靠新进的信息技术和严格的制度 [00:04.000 --> 00:06.400] 为生态文明建设保驾护航 [00:06.400 --> 00:10.000] 让山更绿 天更蓝 水更清 [00:12.000 --> 00:17.000] 这段时间 林业调查员正在对贵阳市新造林进行中期检查 [00:17.000 --> 00:21.000] 过去上山核查要带上罗盘仪和一大堆图纸 [00:21.000 --> 00:23.000] 现在通过智慧林业云平台 [00:23.000...

question

In light of https://github.com/openai/whisper/commit/4179ed2475cc84cba66868b516232ef1b74dacdf, might be worth adding a new GGML.

enhancement

This was blocking me to create `.so` lib using whisper.cpp. I'm not very familiar with CMAKE. Is there another way to fix this without change?

It would be nice to download ready binaries from [release page](https://github.com/ggerganov/whisper.cpp/releases), and not try to build the sources

enhancement
low priority

As for issue #209, this works. We rely on streamlink library to give us a stream, then we proceed similarly to the radio livestream example. I also didn't find a...