Can speech recognition be done in ML.NET?

Question

I am new to Machine Learning. I came across ML.NET (Machine Learning framework by Microsoft). Can we use this framework for speech recognition? What all things you should know if you want to do a speech recognition in ML.NET?

ML.NET does not process audio at this point, just text and images. You can ask for it in the ML.NET github page. — amy8374, Feb 25 '19 at 21:24

score 4 · Accepted Answer · answered Mar 08 '19 at 07:17

4

IMHO, for beginners, modifying sample neural networks/deep learning solutions is a good start point. And, for neural networks the start point is Tensorflow. With ML.net you can consume Tensorflow models.

Audio recognition with Tensorflow:

https://www.tensorflow.org/tutorials/sequences/audio_recognition

A sample for using Tensorflow models with ML.net:

https://github.com/dotnet/machinelearning-samples/tree/master/samples/csharp/getting-started/DeepLearning_ImageClassification_TensorFlow

answered Mar 08 '19 at 07:17

Cihan Yakar

2,402
28
30

1

Thanks Cihan Yakar for suggestion. Will try this. – Kashyap Neeraj Mar 08 '19 at 11:49
1

@KashyapNeeraj – Cihan Yakar Mar 16 '19 at 20:02
Can you share exsample of using wave instead image? – dor Dec 01 '19 at 18:48

Can speech recognition be done in ML.NET?

1 Answers1