nodejs-whisper
nodejs-whisper copied to clipboard
ChetanXpro
Too much is being logged
Verbose flag is set to false, yet too much is logged, like that:
[dev:server]
[dev:server] stderr--- whisper_init_from_file_with_params_no_state: loading model from './models/ggml-base.bin'
[dev:server] whisper_model_load: loading model
[dev:server] whisper_model_load: n_vocab = 51865
[dev:server] whisper_model_load: n_audio_ctx = 1500
[dev:server] whisper_model_load: n_audio_state = 512
[dev:server] whisper_model_load: n_audio_head = 8
[dev:server] whisper_model_load: n_audio_layer = 6
[dev:server] whisper_model_load: n_text_ctx = 448
[dev:server] whisper_model_load: n_text_state = 512
[dev:server] whisper_model_load: n_text_head = 8
[dev:server] whisper_model_load: n_text_layer = 6
[dev:server] whisper_model_load: n_mels = 80
[dev:server] whisper_model_load: ftype = 1
[dev:server] whisper_model_load: qntvr = 0
[dev:server] whisper_model_load: type = 2 (base)
[dev:server] whisper_model_load: adding 1608 extra tokens
Probably from Python/C dunno ...
@binarykitchen Yeah, I was a bit busy for some weeks. Maybe I will be free this week. Feel free to raise a PR, I am happy to merge that.
Cool, new PR is up. I haven't tested it yet, please review carefully: https://github.com/ChetanXpro/nodejs-whisper/pull/118
@ChetanXpro Done, tested. Please review and merge. And test yourself as well.
@binarykitchen Thanks for the PR and sorry for the delayed response! This looks good to me and solves the logging issues. I'll merge this. I'm planning to make some follow-up improvements later to add log level support and control C++ logging (maybe we can do that just by passing verbose mode in cpp command).