ben.carrasco/jambonz-feature-server
1
0
Fork 0
mirror of https://github.com/jambonz/jambonz-feature-server.git synced 2025-12-19 04:17:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

#1088 ignore UtteranceEnd if we have unprocessed words (#1089)

Browse Source 
* #1088 ignore UtteranceEnd if we have unprocessed words

* wip
This commit is contained in:
Dave Horton
2025-02-18 16:30:59 -05:00
committed by Quan HL
parent 6889f0e4ab
commit de1f8e7b1a
1 changed files with 24 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
lib/tasks/gather.js
View File

@@ -757,10 +757,16 @@ class TaskGather extends SttTask {
this.logger.debug('Gather:_onTranscription - got UtteranceEnd event from deepgram but no buffered transcripts');
}
else {
this.logger.debug('Gather:_onTranscription - got UtteranceEnd event from deepgram, return buffered transcript');
evt = this.consolidateTranscripts(this._bufferedTranscripts, 1, this.language, this.vendor);
this._bufferedTranscripts = [];
this._resolve('speech', evt);
const utteranceTime = evt.last_word_end;
if (utteranceTime && this._dgTimeOfLastUnprocessedWord && utteranceTime < this._dgTimeOfLastUnprocessedWord) {
this.logger.debug('Gather:_onTranscription - got UtteranceEnd with unprocessed words, continue listening');
}
else {
this.logger.debug('Gather:_onTranscription - got UtteranceEnd from deepgram, return buffered transcript');
evt = this.consolidateTranscripts(this._bufferedTranscripts, 1, this.language, this.vendor);
this._bufferedTranscripts = [];
this._resolve('speech', evt);
}
}
return;
}
@@ -893,8 +899,21 @@ class TaskGather extends SttTask {
if (originalEvent.is_final && evt.alternatives[0].transcript !== '') {
this.logger.debug({evt}, 'Gather:_onTranscription - buffering a completed (partial) deepgram transcript');
this._bufferedTranscripts.push(evt);
this._dgTimeOfLastUnprocessedWord = null;
}
if (evt.alternatives[0].transcript === '') {
emptyTranscript = true;
}
else if (!originalEvent.is_final) {
/* Deepgram: we have unprocessed words-save last word end time so we can later compare to UtteranceEnd */
const words = originalEvent.channel.alternatives[0].words;
if (words?.length > 0) {
this._dgTimeOfLastUnprocessedWord = words.slice(-1)[0].end;
this.logger.debug(
`TaskGather:_onTranscription - saving word end time: ${this._dgTimeOfLastUnprocessedWord}`);
}
}
if (evt.alternatives[0].transcript === '') emptyTranscript = true;
}
if (!emptyTranscript) {