1 / 15

Pei- ning Chen NTNU CSIE SLP Lab

Discriminative Training Based On An Integrated View Of MPE And MMI In Margin And Error Space Erik McDermott, Shinji Watanabe and Atsushi Nakamura ICASSP 2010. Pei- ning Chen NTNU CSIE SLP Lab. Outline. Introduction Margin-based MPE, MMI, and dMMI

skyla
Download Presentation

Pei- ning Chen NTNU CSIE SLP Lab

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Discriminative Training Based On An Integrated View Of MPE And MMI In Margin And Error SpaceErik McDermott, Shinji Watanabe and Atsushi NakamuraICASSP 2010 Pei-ning Chen NTNU CSIE SLP Lab

  2. Outline • Introduction • Margin-based • MPE, MMI, and dMMI • Macroscopic analysis using the error-indexed forward-backward algorithm • Experimental results • Conclusions

  3. Introduction • It was shown that MPE or MPFE (Minimum Phone Frame Error) corresponds to the derivative of the margin-modified MMI objective function with respect to the margin term. • A new framework, “differenced MMI” (dMMI), was proposed in which the objective function is an integral of MPE-style loss over a given margin interval.

  4. Margin-based MPE Rewrite the cost function in terms of pair-wise comparisons Then the modified MPE loss can be expressed as

  5. Margin-based MMI • Using the same pair-wise comparisons

  6. It is easy to show that MPE (margin-based or not) is the derivative of margin-based MMI with respect to σ

  7. Differenced MMI • It is defined in terms of an integral of MPE loss over a given margin interval

  8. Optimization based on dMMI • For a given arc q in a recognition lattice for utterance Xr, • where is the standard arc posterior probability or occupancy calculated with the Forward-Backward algorithm. • The corresponding lattice arc occupancies are subtracted and divided by σ2 − σ1:

  9. Optimization based on dMMI • The total gradient for all parameter components Λi, summed over all training utterances and all Qr arcs in each utterance’s recognition lattice, can then be calculated

  10. The error-indexed forward-backward algorithm • An aggregate probability mass for all lattice strings with the same total error count j : • The corresponding margin-modified error group occupancy is

  11. The standard (σ = 0) error group MPE derivative is • The aggregated dMMIderivative is

  12. 454

  13. Experimental results

  14. Conclusion • A new approach for DT, “differenced MMI”. • Experiments confirmed that a close approximation to MPE can be implemented using dMMI. • Aggregate error-group statistics show that the choice of interval affects the relative weighting of different error levels during training. • The proper choice of margin interval is a topic for future research.

More Related