bobby/open_dbm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add documentation based on docusarus for opendbm

Browse Source
This commit is contained in:
Rudy Haryanto
2022-10-04 03:34:47 +07:00
parent a574bc6870
commit c6bc952569
62 changed files with 56077 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/docs/speech.md Normal file
View File

@@ -0,0 +1,8 @@
---
id: speech
title: Speech
---
The previous section focused on acoustic properties of speech. This section measures characteristics of language. First, any speech detected in the audio file is transcribed into text using [DeepSpeech](https://github.com/mozilla/DeepSpeech), an open-source speech transcription tool. From the transcribed text, a number of open source natural language processing tools are used to derive characteristics of speech.
Speech biomarkers are slightly different from the other categories of digital biomarkers as the ‘raw variables’ are really just the transcribed speech (on how to acquire the transcribed speech, see Section 3.1.4). For reasons that are unimportant, we do still calculate raw variables and derived variables separately here, but there is absolutely no difference between the raw and derived variables. Hence, we recommend conducting any analysis on speech data simply from the derived variables, listed below.