Brandon Lee

yea they are .wav with monochannel 16000 sample rates I generated them with [librosa write file](http://man.hubwiz.com/docset/LibROSA.docset/Contents/Resources/Documents/generated/librosa.output.write_wav.html#librosa.output.write_wav) and for some datasets I generated, it works fine. It just randomly happens that...

Things still fail with the correct format ``` import soundfile as sf data = (item.audio_data.numpy() * 32767).astype('int16') sf.write(output_path, data, 16000, subtype='PCM_16', endian='LITTLE') ```

I have debugged the issue. [pyache](https://github.com/MycroftAI/pyache) was getting into the way. I think others fall into the same trap. is it possible for u to add support for disabling the...

I would first make sure `VOCAB` and `INFERENCE_SEQUENCE` are correct. let me see if I can clean things up and provide pretrained model

please refer to https://github.com/castorini/howl/pull/39 for the pretrained model. You might need to append appropriate env vars for your command. ``` VOCAB='[" hey","fire","fox"]' INFERENCE_SEQUENCE=[0,1,2] INFERENCE_THRESHOLD=0 NUM_MELS=40 MAX_WINDOW_SIZE_SECONDS=0.5 python -m howl.run.demo --model...

This is the workspace for hey firefox, https://github.com/castorini/howl-models/tree/master/howl/hey-fire-fox

I would be pretty surprised if [this instruction](https://github.com/castorini/howl#hey-firefox) doesn't work. Did you download the datasets in the read me before it was broken?

I see @daemon can you check the server? also how were these dataset constructed, same as in this [instruction](https://github.com/castorini/howl#preparing-a-dataset)?

did you also do the following steps? alignment? I guess you have. because without doing so, you won't be able to feed it into the howl

This should fix the broken link #43