I want to transcribe telephonic conversation audio files(8Khz). Other that resampling it to 16K what is the best option.
I don’t think I’ve understood your question. Do you want to directly input the model with audios of 8kHz instead of resampling?
Yeah. Something like that.
Hi Sumeet,
Have you been able to transcribe telephonic conversations? If so, I presume you had to train an acoustic model for telephone audio? And if so, would you be willing to share it?
Thx.