Daniel Brower

That's odd that the raw file is 44100 Hz. I specified it to be 16000 Hz.

At TLSphinx: Decoder, it says that the input file needs to be unheadered. I think that means it needs to be a file of type raw, with the ".raw" extension. However, at Q: How can I decode audio encoded with a codec (mp3, mu-law, mp4, g729), it shows that the input file needs to be a wav file. Can the input file be unheadered if it's a wav file? I have tried using both a wav file and a raw file, but either way I get whitespace only in the text of the Hypothesis. I followed the instructions at Q: How...

Hypothesis.text empty or whitespace

recognize words in audio file without audio playing