search

mozilla / DeepSpeech

DeepSpeech is an open source embedded (offline, on-device) speech-to-text engine which can run in real time on devices ranging from a Raspberry Pi 4 to high power GPU servers.
Mozilla Public License 2.0
25.15k stars 3.95k forks source link

mfcc vs filterbanks #501

Closed ghost closed 7 years ago

ghost commented 7 years ago

Hey @reuben and mozilla team, in the original Baidu DeepSpeech-1 paper, the input is filterbanks instead of mfcc. Also you had pushed some changes to use that. Have you compared the results of both on any dataset like Switchboard. Any visible insight will be great

reuben commented 7 years ago

We have not compared MFSC vs MFCC yet, mostly due to all the different experiments we want to do competing for server time. Issue #353 is open for exactly this question.

lock[bot] commented 5 years ago

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.