Smart audio playback
Embodiments are directed to controlling playback of recordings. The recording can comprise an audio recording, audio/visual recording, voicemail message, or other recording having an audio component. According to one embodiment, a method can comprise capturing an audio recording of speech of at least one person and determining, a context for each of a plurality of portions of the audio recording based on natural language processing of the audio recording. One or more transition points between the portions of the audio recording can be identified. Each transition point can indicate a change in the determined context between the portions. A playback interface providing a representation of the audio recording and each of the identified transition points can be presented and the audio recording can be played based on input received through the playback interface.
1. A method for controlling playback of audio recordings, the method comprising:
capturing, by a processor of a media processing system, an audio recording of speech of at least one person;
determining, by the processor of the media media processing system, one or more keywords and a context for each of a plurality of portions of the audio recording based on natural language processing of the audio recording;
identifying, by the processor of the media media processing system, one or more transition points between the plurality of portions of the audio recording, each transition point indicating a change in the determined context between the plurality of portions;
presenting, by the media processing system, a playback interface providing a representation of the audio recording and each of the identified one or more transition points;
playing, by the media processing system, the audio recording based on input received through the playback interface; and
updating a user profile associated with a user of the playback interface based on selection of a specific transition point and one or more associated keywords of the specific transition point, wherein the user profile comprises a set of ranked keywords of interest to the user, and wherein updating the user profile comprises re-ranking the set of ranked keywords of interest to the user based on the one or more associated keywords of the selected specific transition point.
2. The method of claim 1 , wherein determining the context for each of the plurality of portions of the audio recording comprises transcribing the audio recording into a text transcript, annotating the text transcript based on natural language processing of the text transcript, and deducing the context for each of the plurality of portions of the audio recording based on the annotated text transcript.
3. The method of claim 1 , wherein determining the context for each of the plurality of portions of the audio recording comprises annotating the audio recording based on natural language processing of the audio recording and deducing the context for each of the plurality of portions of the audio recording based on the annotated audio recording.
4. The method of claim 1 , wherein the playback interface comprises a graphical timeline element representing a length of the audio recording, the graphical timeline element divided to represent the plurality of portions of the audio recording by one or more graphical transition markers positioned along the graphical timeline element, each graphical transition marker corresponding to one of the identified one or more transition points and indicating a point in time within the audio recording for the corresponding one of the identified one or more transition points, at least one textual context indicator for each of the plurality of portions of the audio recording, and one or more textual keyword indicators for each of the plurality of portions of the audio recording.
5. The method of claim 4 , wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to the selected specific transition point.
6. The method of claim 1 , further comprising reading the user profile for the user of the playback interface, the user profile defining one or more contexts of interest to the user, wherein the input received through the playback interface indicates a request to fast forward, and wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to a different transition point, the different transition point being selected based on the one or more contexts of interest to the user.
7. A system comprising:
a processor; and
a memory coupled with and readable by the processor and storing therein a set of instructions which, when executed by the processor, causes the processor to control playback of audio recordings by:
capturing an audio recording of speech of at least one person;
determining one or more keywords and a context for each of a plurality of portions of the audio recording based on natural language processing of the audio recording;
identifying one or more transition points between the plurality of portions of the audio recording, each transition point indicating a change in the determined context between the plurality of portions;
presenting a playback interface providing a representation of the audio recording and each of the identified one or more transition points;
playing the audio recording based on input received through the playback interface; and
updating a user profile associated with a user of the playback interface based on selection of a specific transition point and one or more associated keywords of the specific transition point, wherein the user profile comprises a set of ranked keywords of interest to the user, and wherein updating the user profile comprises re-ranking the set of ranked keywords of interest to the user based on the one or more associated keywords of the selected specific transition point.
8. The system of claim 7 , wherein determining the context for each of the plurality of portions of the audio recording comprises:
transcribing the audio recording into a text transcript;
annotating the text transcript based on natural language processing of the text transcript; and
deducing the context for each of the plurality of portions of the audio recording based on the annotated text transcript.
9. The system of claim 7 , wherein determining the context for each of the plurality of portions of the audio recording comprises annotating the audio recording based on natural language processing of the audio recording and deducing the context for each of the plurality of portions of the audio recording based on the annotated audio recording.
10. The system of claim 7 , wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to the selected specific transition point.
11. The system of claim 7 , wherein the set of instructions further cause the processor to read the user profile for the user of the playback interface, the user profile defining one or more contexts of interest to the user.
12. The system of claim 11 , wherein the input received through the playback interface indicates a request to fast forward, and wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to a different transition point, the different transition point being selected based on the one or more contexts of interest to the user.
13. A non-transitory, computer-readable medium comprising a set of instructions stored therein which, when executed by a processor, causes the processor to control playback of audio recordings by:
capturing an audio recording of speech of at least one person;
determining one or more keywords and a context for each of a plurality of portions of the audio recording based on natural language processing of the audio recording;
identifying one or more transition points between the plurality of portions of the audio recording, each transition point indicating a change in the determined context between the plurality of portions;
presenting a playback interface providing a representation of the audio recording and each of the identified one or more transition points;
playing the audio recording based on input received through the playback interface; and
updating a user profile associated with a user of the playback interface based on selection of a specific transition point and one or more associated keywords of the specific transition point, wherein the user profile comprises a set of ranked keywords of interest to the user, and wherein updating the user profile comprises re-ranking the set of ranked keywords of interest to the user based on the one or more associated keywords of the selected specific transition point.
14. The non-transitory, computer-readable medium of claim 13 , wherein determining the context for each of the plurality of portions of the audio recording comprises:
transcribing the audio recording into a text transcript;
annotating the text transcript based on natural language processing of the text transcript; and
deducing the context for each of the plurality of portions of the audio recording based on the annotated text transcript.
15. The non-transitory, computer-readable medium of claim 13 , wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to the selected specific transition point.
16. The non-transitory, computer-readable medium of claim 15 , wherein the set of instructions further cause the processor to read the user profile for the user of the playback interface, the user profile defining one or more contexts of interest to the user.
17. The non-transitory, computer-readable medium of claim 16 , wherein the input received through the playback interface indicates a request to fast forward, and wherein playing the audio recording based on the input received through the playback interface comprises fast forwarding to a different transition point, the different transition point being selected based on the one or more contexts of interest to the user.