O'Reilly logo

Hands-On Natural Language Processing with Python by Rajalingappaa Shanmugamani, Rajesh Arumugam

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

State-of-the-art in speech recognition

The original DeepSpeech architecture uses a language model to correct some errors in the character sequences. It combines both the output of the RNN and a language model to arrive at the most probable text sequence for a given speech recording. As attention-based methods are getting popular due to their effectiveness on improving accuracies over non-attention-based systems, it has been adopted even for speech. The paper Listen, Attend, and Spell: A Neural Network for Large Vocabulary Conversational Speech Recognition (https://ai.google/research/pubs/pub44926) from Google incorporates an attention mechanism to transcribe speech recordings.

The main advantage of this system is that it uses an attention-based ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required