Pengembangan voice recognition menggunakan tensorflow ichenli train_mnist_image untuk aplikasi deep speech

Diana Firdaus, Rahmatika (2018) Pengembangan voice recognition menggunakan tensorflow ichenli train_mnist_image untuk aplikasi deep speech. Project Report. Institut Teknologi Telkom Purwokerto, Perpustakaan Institut Teknologi Telkom Purwokerto. (Unpublished)

	Text COVER.pdf Download (769kB)
Preview	Text ABSTRAK.pdf Download (6kB) | Preview
Preview	Text ABSTRACT.pdf Download (5kB) | Preview
Preview	Text BAB I.pdf Download (149kB) | Preview
	Text BAB II.pdf Restricted to Registered users only Download (20kB) | Request a copy
	Text BAB III.pdf Restricted to Registered users only Download (691kB) | Request a copy
Preview	Text DAFTAR PUSTAKA.pdf Download (81kB) | Preview
Preview	Text LAMPIRAN.pdf Download (24kB) | Preview

Abstract

Practical Work (KP) is a mandatory activity that must be carried out by every student of the Purwokerto Institute of Technology as a requirement to take a thesis course. The author is placed in PT Menara Multimedia Telkom Solution Jakarta. The author is given the task of making a Deep Speech application or voice recognition device. The Deep Speech application aims to convert voice to text automatically to recognize various sound characters according to the conversation content. In a variety of smart phones already embedded many applications that can convert sound into text according to the command. The author is given the task of studying a Tensorflow Ichenli Train_Mnist_Image. The aim of learning Tensorflow is to find the simplest script with the highest accuracy to be implemented into the Deep Speech Application script. The result is the Ichenli Train_Mnist_Image script is not suitable to be implemented because it is long and complicated and very small accuracy of 0.1105

Actions (login required)

View Item