ben.carrasco/jambonz-feature-server
1
0
Fork 0
mirror of https://github.com/jambonz/jambonz-feature-server.git synced 2025-12-20 16:50:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

punctuation for microsoft (#566)

Browse Source 
* punctuation for microsoft

* wip
This commit is contained in:
Hoan Luu Huu
2023-12-18 20:38:05 +07:00
committed by GitHub
parent bcb4bf43bf
commit 30977b309c
3 changed files with 10 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
lib/utils/transcription-utils.js
View File

@@ -338,19 +338,20 @@ const normalizeNuance = (evt, channel, language) => {
};
};
const normalizeMicrosoft = (evt, channel, language) => {
const normalizeMicrosoft = (evt, channel, language, punctuation = true) => {
const copy = JSON.parse(JSON.stringify(evt));
const nbest = evt.NBest;
const language_code = evt.PrimaryLanguage?.Language || language;
const alternatives = nbest ? nbest.map((n) => {
return {
confidence: n.Confidence,
transcript: n.Display
// remove all puntuation if needed
transcript: punctuation ? n.Display : n.Display.replace(/\p{P}/gu, '')
};
}) :
[
{
transcript: evt.DisplayText || evt.Text
transcript: punctuation ? evt.DisplayText || evt.Text : (evt.DisplayText || evt.Text).replace(/\p{P}/gu, '')
}
];
@@ -400,14 +401,14 @@ const normalizeAssemblyAi = (evt, channel, language) => {
};
module.exports = (logger) => {
const normalizeTranscription = (evt, vendor, channel, language, shortUtterance) => {
const normalizeTranscription = (evt, vendor, channel, language, shortUtterance, punctuation) => {
//logger.debug({ evt, vendor, channel, language }, 'normalizeTranscription');
switch (vendor) {
case 'deepgram':
return normalizeDeepgram(evt, channel, language, shortUtterance);
case 'microsoft':
return normalizeMicrosoft(evt, channel, language);
return normalizeMicrosoft(evt, channel, language, punctuation);
case 'google':
return normalizeGoogle(evt, channel, language);
case 'aws':