Train for only one voice

[mankaden]

Hi Newbie here. Been trying to install DeepSpeech for 3 days now, so while i figure that out - Is it possible to train DeepSpeech using only my voice? The idea being get it to 0% word error rate for me only, if that makes any sense…

[daniel.cruzado]

Here is a tutorial which does something quite similar to what you want.

TUTORIAL : How I trained a specific french model to control my robot

Tutorial How to build your homemade deepspeech model from scratch Adapt links and params with your needs… For my robotic project, I needed to create a small monospeaker model, with nearly 1000 sentences orders (not just single word !) I recorded wav’s with a Respeaker Microphone Array : https://www.seeedstudio.com/ReSpeaker-Mic-Array-Far-field-w%2F-7-PDM-Microphones-p-2719.html Wav’s were recorder with the following params : mono / 16 bits / 16 k. The use of the google vad lib helped me t…

[mankaden]

Thanks I actually literally just found this

[elpimous_robot (Vincent Foucault)]

Hi @mankaden,
reaching 0% Wer is utopia ! Even with mono speaker.

But, with patience, enough datas, and same as final target audio files, you could obtain very good results… under 0.05 WER (5%)
Data augmentation helps a lot too !!
Correct learning params too !!
“Bref”, be patient !! LOL (ready to help you…but learn tuto before!!!)

[mankaden]

Someones got to be the first to reach 0%, might as well aim for utopia

Thanks I will take you up on that one day!