Method for voicemail quality detection

1. A computer-implemented method for non-intrusive speech quality detection without a reference signal comprising:

receiving, at a computing device configured to convert voicemail to text, a first speech signal associated with a user;

extracting one or more short-term features from the first speech signal wherein extracting short-term features includes extracting a time frame of between 10-50 ms, wherein the one or more short term features include a Hilbert envelope based feature and a linear predictive coding residual;

determining one or more statistics of each of the one or more short-term features from the first speech signal;

classifying the one or more statistics as belonging to one of a set of quality classes, wherein classifying the one or more statistics includes modeling a speech quality class using a binary tree classifier; and

automatically generating at least one training database, based upon, at least in part, the first speech signal and an intrusive speech quality algorithm, wherein the intrusive speech quality algorithm is not used during the receiving, extracting, determining, and classifying operations.

2. The method of claim 1 , wherein the one or more statistics include at least one of mean, variance, skewness, and kurtosis.

3. The method of claim 1 , wherein the one or more short-term features include at least one of pitch frequency, zero crossing rate, importance weighted signal to noise ratio, and difference from long-term average speech magnitude spectrum features.

4. The method of claim 3 , wherein the difference from long-term average speech magnitude spectrum features includes at least one of flatness, centroid, and a power spectrum of long term deviation.

5. The method of claim 1 , wherein classifying is based upon, at least in part, non-intrusive classification of message quality.

6. The method of claim 5 , wherein classifying is performed per each time frame.

7. The method of claim 1 , further comprising:

extracting one or more long-term features from the first speech signal.

8. The method of claim 7 , wherein the one or more long-term features includes a percentage of energy per frequency band.

9. A non-transitory computer-readable storage medium having stored thereon instructions, which when executed by a processor result in one or more operations for non-intrusive speech quality detection without a reference signal, the operations comprising:

receiving, at a computing device configured to convert voicemail to text, a first speech signal associated with a particular user;

extracting one or more short-term features from the first speech signal wherein extracting short-term features includes extracting a time frame of between 10-50 ms, wherein the one or more short term features include a Hilbert envelope based feature and a linear predictive coding residual;

determining one or more statistics of each of the one or more short-term features from the first speech signal;

classifying the one or more statistics as belonging to one of a set of quality classes, wherein classifying the one or more statistics includes modeling a speech quality class using a binary tree classifier; and

automatically generating at least one training database, based upon, at least in part, the first speech signal and an intrusive speech quality algorithm, wherein the intrusive speech quality algorithm is not used during the receiving, extracting, determining, and classifying operations.

10. The non-transitory computer-readable medium of claim 9 , wherein the one or more statistics include at least one of mean, variance, skewness, and kurtosis.

11. The non-transitory computer-readable medium of claim 9 , wherein the one or more short-term features include at least one of pitch frequency, zero crossing rate, importance weighted signal to noise ratio, and difference from long-term average speech magnitude spectrum features.

12. The non-transitory computer-readable medium of claim 9 , wherein classifying is based upon, at least in part, non-intrusive classification of message quality.

13. The non-transitory computer-readable medium of claim 12 , wherein classifying is performed per each time frame.

14. A voicemail to text system configured to perform non-intrusive speech quality detection without a reference signal comprising:

one or more processors configured to receive a first speech signal associated with a particular user, the one or more processors further configured to extract one or more short-term features from the first speech signal wherein extracting short-term features includes extracting a time frame of between 10-50 ms, wherein the one or more short term features include a Hilbert envelope based feature and a linear predictive coding residual, the one or more processors further configured to determine one or more statistics of each of the one or more short-term features from the first speech signal, the one or more processors further configured to classify the one or more statistics as belonging to one of a set of quality classes, wherein classifying the one or more statistics includes modeling a speech quality class using a binary tree classifier, the one or more processors further configured to automatically generate at least one training database, based upon, at least in part, the first speech signal and an intrusive speech quality algorithm, wherein the intrusive speech quality algorithm is not used during the receiving, extracting, determining, and classifying operations.

15. The system of claim 14 , wherein the one or more statistics include at least one of mean, variance, skewness, and kurtosis.

16. The system of claim 14 , wherein the one or more short-term features include at least one of pitch frequency, zero crossing rate, importance weighted signal to noise ratio, and difference from long-term average speech magnitude spectrum features.

17. The system of claim 14 , wherein classifying is based upon, at least in part, non-intrusive classification of speech quality.