es-ES Not show phoneme

[hiddow]

Is your feature request related to a problem? Please describe. Currently, Azure Speech Assessment provides valuable insights into pronunciation accuracy, but it can be challenging for language learners focusing on Spanish (es-ES) to pinpoint the exact phonemes where they may be struggling. Detailed phoneme-level feedback would significantly enhance the learning process.

Describe the solution you'd like I would like Azure Speech Assessment to include a detailed breakdown of the output at the phoneme level for Spanish (es-ES) assessments. This would involve:

Individual Phoneme Scores: Assign accuracy ratings (e.g., Correct, Substitution, Deletion, Insertion) to each phoneme within the assessed text. Visual Representation: Provide a clear visual representation of this breakdown, potentially highlighting problematic phonemes for quick identification.

Describe alternatives you've considered Manual Phonetic Analysis: It's possible to manually transcribe the audio using the International Phonetic Alphabet (IPA) and compare against a reference, but this is extremely time-consuming and requires specialized knowledge. Third-Party Tools: There are some limited phonetic analysis tools, but integrating this functionality within Azure's suite would be far more streamlined.

Additional context Detailed phoneme-level feedback has the potential to greatly benefit:

Spanish Language Learners: Allows users to target specific areas of pronunciation difficulty, leading to more focused and effective practice. Language Instructors: Provides actionable data for lesson planning and individualized pronunciation guidance.

[jpalvarezl]

Hello @hiddow thank you for using the SDK. As you may be aware currently es-ES is not part of the supported languages for this feature.

I will notify the feature team for this area to review the issue and they will evaluate, given their current roadmap and resources, whether this can be moved into active development.

[yulin-li]

@wangkenpu could you check?

[wangkenpu]

It's by design. For new feature requests, please fill Microsoft Forms. After you submit the form, our team will contact you in 2 weeks.

[github-actions[bot]]

This item has been open without activity for 19 days. Provide a comment on status and remove "update needed" label.

[pankopon]

Closed as by design.

[fabswt]

People really like getting pronunciation feedback on Gliglish.com for en-US (I'm using Azure's Pronunciation Assessment), but I cannot extend it to more languages without phoneme-level data.

It's a little absurd to offer phoneme-level scores without phoneme-level data.

I submitted a response to the form mentioned above.