How do I use speech recognition to transcribe from a recording?

[owl8000]

I want to know how to use a digital recording of a spoken presentation as
input for the speech recognition dictating. I can see how to use speech
recognition to dictate through a microphone. So there must be a way to use a
recording.

 

[grammatim]

Can you simply run a cable from the audio output to the microphone
input?

 

[Peter A]

Can you simply run a cable from the audio output to the microphone
input?

No, audio output is typically at a much higher voltage than mic output.
In theory one could run a cable from the audio output to the line in
jack, but I don't know whether the speech recognition software
"listens" on that line.

For what it's worth, speech recognition software does not work worth a
damn until it has been trained with a specific speaker's voice and
speech habits. You cannot expect it to do well otherwise.

 

[grammatim]

No, audio output is typically at a much higher voltage than mic output.
In theory one could run a cable from the audio output to the line in
jack, but I don't know whether the speech recognition software
"listens" on that line.

For what it's worth, speech recognition software does not work worth a
damn until it has been trained with a specific speaker's voice and
speech habits. You cannot expect it to do well otherwise.

As a linguist, I'd be inclined to say it doesn't work at all.