wit-ai / wit-ai/pywit

Request for getting alternative matches based on confidence for STT

Open
#152 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.5k
Forks
353
PR merge metrics
No merged PRs in 30d

Description

Do you want to request a feature, report a bug, or ask a question about pywit?

Yes, I couldn't find a feature on WIT to get the next best alternatives to the STT (Speech-to-text) predictions.

What is the current behavior?

Currently, during Speech-to-Text we get the phrase that has the highest confidence according to WIT and the training we have done. But, we do not have the possibility to get the nearest alternatives to that spoken phrase. For eg. check

This could be particularly useful for:

  • while making chatbots and asking the user for confirmation on something they said
  • For correcting the prediction of a word while converting to text based on context of the sentence.
  • for rare words that isn't accurately predicted by WIT.

What is the expected behavior?

would it be possible to get 2-5 of the next best possible interpretations that the model makes from the audio file?

If applicable, what is the App ID where you are experiencing this issue? If you do not provide this, we cannot help.

745320612840161

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No file or test is named in the issue. Start by tracing pywit's speech-to-text request and response handling, then verify whether Wit.ai exposes ranked alternatives and confidence values. Done means establishing whether 2–5 alternatives can be returned and documenting or implementing the supported behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
api
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.