Could you tell me why the recognized text is not displayed in the console?
First, avoid screenshots. Seconds, it’s not nothing, it’s empty. That depends on what you have in your audio file … What model you are using … ?
I understood about the screenshots. I’m using default settings and model v0.4.1 from this repo https://github.com/mozilla/DeepSpeech#getting-the-pre-trained-model. And wav file https://drive.google.com/file/d/19xnT2gdXvtGhT8ELz6xo9dSJLuBMLkJb/view?usp=sharing
I see 8k in the filename, can you share more informations? I can’t access your google drive link.
file has been converted to 16k format, its only name. Try one more time open link https://drive.google.com/open?id=19xnT2gdXvtGhT8ELz6xo9dSJLuBMLkJb
Please, can’t you just share the infos about the file? rate, channels, etc. ?
Its just file from http://www.voiptroubleshooter.com/open_speech/american.html and converted to 16k with this sservice https://www.conversion-tool.com/wav/
Thanks for your help, is it really that hard that you just share the informations of your file ? Should I really have to redo the whole conversion, not being sure if I have exactly the same, not controlling the parameters ?
@lissyx I made mistake at –alphabet models/alphabet.txt using alphabt.txt instead alphabet.txt. Thak you for help.
1 Like
So now you have proper inference computed? Is this models/alphabt.txt really existing ? It should not have worked at all.