System and method for automatically transcribing voicemail
View Patent ↗Disclosed herein are systems, computer-implemented methods, and tangible computer-readable media for automatically transcribing voicemail. The method includes receiving a plurality of voicemail messages from callers, identifying for each voicemail message in the plurality of voicemail messages a first frequency with which the respective caller leaves voicemails, identifying for each voicemail message in the plurality of voicemail messages a second frequency with which a user requests transcription of each voicemail, assigning a priority ranking to each voicemail message in the plurality of voicemail messages based on the respective first frequency and the respective second frequency, and transcribing untranscribed voicemail messages with a highest priority ranking. The method can include establishing a priority ranking threshold and repeatedly transcribing a next highest ranking untranscribed voicemail message until no further untranscribed voicemail messages remain above the priority ranking threshold.
1. A method comprising:
receiving a plurality of voicemail messages;
assigning, via a processor, a priority to each voicemail message in the plurality of voicemail messages, the priority based on a call frequency by a caller associated with each voicemail message and a user transcription request frequency, to yield prioritized untranscribed voicemail messages;
determining a highest ranked priority voicemail in the prioritized untranscribed voicemail messages; and
transcribing the highest ranked priority voicemail.
2. The method of claim 1 , further comprising repeatedly transcribing a next highest priority untranscribed voicemail message until no further untranscribed voicemail messages remain.
3. The method of claim 1 , further comprising:
establishing a priority ranking threshold; and
repeatedly transcribing a next highest priority voicemail message until no further untranscribed voicemail messages remain above the priority ranking threshold.
4. The method of claim 1 , further comprising building a voice model of a respective caller while transcribing voicemail messages.
5. The method of claim 1 , wherein assigning the priority to each voicemail message is further based on a user profile.
6. The method of claim 1 , further comprising generating speaker-specific syllable libraries while transcribing the untranscribed voicemail messages.
7. The method of claim 6 , further comprising:
receiving a text-based message from a speaker associated with a generated speaker-specific syllable library; and
generating an audible version of the text-based message based on the generated speaker-specific syllable library.
8. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving a plurality of voicemail messages;
assigning a priority to each voicemail message in the plurality of voicemail messages, the priority based on a call frequency by a caller associated with each voicemail message and a user transcription request frequency, to yield prioritized untranscribed voicemail messages;
determining a highest ranked priority voicemail in the prioritized untranscribed voicemail messages; and
transcribing the highest ranked priority voicemail.
9. The system of claim 8 , the computer-readable storage medium having additional instructions stored which result in operations comprising repeatedly transcribing a next highest priority untranscribed voicemail message until no further untranscribed voicemail messages remain.
10. The system of claim 8 , the computer-readable storage medium having additional instructions stored which result in operations comprising:
establishing a priority ranking threshold; and
repeatedly transcribing a next highest priority voicemail message until no further untranscribed voicemail messages remain above the priority ranking threshold.
11. The system of claim 8 , the computer-readable storage medium having additional instructions stored which in operations comprising building a voice model of a respective caller while transcribing voicemail messages.
12. The system of claim 8 , wherein assigning the priority to each voicemail message is further based on a user profile.
13. The system of claim 8 , wherein the computer-readable storage medium having additional instructions stored which result in operations comprising generating speaker-specific syllable libraries while transcribing the untranscribed voicemail messages.
14. The system of claim 13 , the computer-readable storage medium having additional instructions stored which result in operations comprising:
receiving a text-based message from a speaker associated with a generated speaker-specific syllable library; and
generating an audible version of the text-based message based at least in part on the generated speaker-specific syllable library.
15. A computer-readable storage device having instructions stored which, when executed by a processor, cause the processor to perform operations comprising:
receiving a plurality of voicemail messages;
assigning a priority to each voicemail message in the plurality of voicemail messages, the priority based on a call frequency by a caller associated with each voicemail message and a user transcription request frequency, to yield prioritized untranscribed voicemail messages;
determining a highest ranked priority voicemail in the prioritized untranscribed voicemail messages; and
transcribing the highest ranked priority voicemail.
16. The computer-readable storage device of claim 15 , having additional instructions which result in operations comprising repeatedly transcribing a next highest priority untranscribed voicemail message until no further untranscribed voicemail messages remain.
17. The computer-readable storage device of claim 15 , having additional instructions which result in operations comprising:
establishing a priority ranking threshold; and
repeatedly transcribing a next highest priority voicemail message until no further untranscribed voicemail messages remain above the priority ranking threshold.
18. The computer-readable storage device of claim 15 , having additional instructions which result in operations comprising building a voice model of a respective caller while transcribing voicemail messages.