You are here

CloudASR: A Web Platform for Automatic Speech Recognition

Event date: 
Thursday, 30 July, 2015 - 16:00 to 17:00
Sala Direzione
Ondřej Klejch

Final presentation of summer internship student.

In this talk I will present a cloud platform for automatic speech recognition, CloudASR, built on top of Kaldi speech recognition toolkit. The platform supports both batch and online speech recognition mode and it has an annotation interface for transcription of the submitted recordings. The key features of the platform are scalability, customizability and easy deployment. Benchmarks of the platform show that the platform achieves comparable performance with Google Speech API in terms of latency and it can achieve better accuracy on limited domains. Furthermore, the benchmarks show that the platform is able to handle more than 1000 parallel requests given enough computational resources.
In this talk I will describe motivation, design choices and technologies that allowed us to build CloudASR.