Johanes Effendi

Thank you very much @david-waterworth , I got the same error and it solves my problem! I'm on MacOS Catalina.

Thank you @webbp for your solution, it works! Just want to add @backspacetg note, in the case you install `fftw3` by source, flashlight will look for `libfftw3.so.3` in the fftw...

Bump. got similar error. Both `target_dict` and `eval_wer` does not make sense to be in Audio**Pretraining**Config, as pretraining does not use text. Has someone found the solution? ``` omegaconf.errors.ConfigKeyError: Key...

@keunwoochoi `eval_wer` is not available in pretraining task, override the task into finetuning, like this: ``` model_override_rules = ast.literal_eval(args.model_overrides) model_override_rules['task'] = {'_name': 'audio_finetuning'} models, saved_cfg, task = checkpoint_utils.load_model_ensemble_and_task( arg_overrides=model_override_rules, ........

@kylebgorman thank you for your hard work! I agree with @lextoumbourou and @david-waterworth, if it's not supported by Apple anymore, I think it's reasonable not to support it anymore. BTW...