add confidence key to the MbrResult method in Kaldi Recognizer #563
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
sskorol
reviewed
May 30, 2021
| @@ -423,6 +424,7 @@ const char *KaldiRecognizer::MbrResult(CompactLattice &clat) | |||
| word["start"] = samples_round_start_ / sample_frequency_ + (frame_offset_ + times[i].first) * 0.03; | |||
| word["end"] = samples_round_start_ / sample_frequency_ + (frame_offset_ + times[i].second) * 0.03; | |||
| word["conf"] = conf[i]; | |||
| confidence += conf[i]; | |||
There was a problem hiding this comment.
I don't believe it should work this way. Assuming a confidence is a value in [0;1] range, you can't just sum all the values to get an overall confidence.
|
I don't think this is the right way to calculate confidence. Why average and not minimum for example? I need to think more about it. |
|
We need the confidence value to make some decision within our application. The average gives a better approximation of the overall confidence than max/min. We also compare the converted text with the list of accepted words or sentences using levenshtein distance method. Using these two values we will make a decision. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Our IVR system is built on Freeswitch + Rayo + Adhearsion. We are using Unimrcp with vosk-api and Kaldi to perform Speech to text for our on premise solution. We observed that the overall confidence of the conversion is not coming as part of the
MbrResultmethod. We are expecting aconfidencekey as part of the result so that the Adhearsion application will make a decision. I have added that as part of this PR.