Text-to-speech from media content item snippets
A text-to-speech engine creates audio output that includes synthesized speech and one or more media content item snippets. The input text is obtained and partitioned into text sets. A track having lyrics that match a part of one of the text sets is identified. The location of the track's audio that contains the lyric is extracted based on forced alignment data. The extracted audio is combined with synthesized speech corresponding to the remainder of the input text to form audio output.
1. A method for providing text-to-speech functionality, comprising:
splitting an input text into a first text set and a second text set, wherein the first text set is selected for the splitting based on the first text set being associated with suitable lyrics or a suitable lyrical characteristic matched by a key value pair in a dictionary structure, and wherein the second text set is selected for the splitting based on the second text set being suitable for speech synthesis;
extracting an audio snippet from a track corresponding to the first text set, wherein the audio snippet is a portion of the track that begins at a start time within the track and ends at an end time within the track;
creating a synthesized utterance of at least one word of the second text set;
concatenating the audio snippet and the synthesized utterance to form combined audio; and
providing an audio output that includes the combined audio.
2. The method of claim 1 , wherein the track is selected randomly.
3. The method of claim 1 , wherein the track is selected by presenting, via a user device interface, a prompt to a user to select the track from a plurality of tracks.
4. The method of claim 1 , wherein the track is selected from a plurality of tracks by:
sorting the plurality of tracks based on a quality associated with the first text set into a sorted list; and
selecting the track based on the sorted list.
5. The method of claim 1 , wherein the track is selected from a plurality of tracks by:
sorting the plurality of tracks based on a taste profile of a user into a sorted list; and
selecting the track based on the sorted list.
6. The method of claim 1 , wherein the track is selected based on a play list associated with the input text.
7. The method of claim 1 , wherein the track is selected from a plurality of tracks by:
determining a suitability of each track in the plurality of tracks, including by determining how clearly a lyric can be heard; and
selecting the track based on the suitability.
8. The method of claim 1 , wherein the audio snippet contains one or more words of the first text set and, and wherein the combined audio contains the one or more words of the first text set and the at least one word of the second text set.
9. The method of claim 1 , wherein splitting the input text into the first text set and the second text set comprises splitting the input text into the first text set, the second text set, and a third text set, the method further comprising:
extracting a second audio snippet from a second track based on the third text set, wherein the combined audio further includes the second audio snippet.
10. A system providing text-to-speech functionality, comprising a text-to-speech engine configured to:
split an input text into a first text set and a second text set, wherein the first text set is selected for the splitting based on the first text set being associated with suitable lyrics or a suitable lyrical characteristic matched by a key value pair in a dictionary structure, and wherein the second text set is selected for the splitting based on the second text set being suitable for speech synthesis;
extract an audio snippet from a track corresponding to the first text set, wherein the audio snippet is a portion of the track that begins at a start time within the track and ends at an end time within the track;
create a synthesized utterance of at least one word of the second text set;
concatenate the audio snippet and the synthesized utterance to form combined audio; and
provide an audio output that includes the combined audio.
11. The system of claim 10 , wherein the track is selected randomly.
12. The system of claim 10 , wherein the track is selected by presenting, via a user device interface, a prompt to a user to select the track from a plurality of tracks.
13. The system of claim 10 , wherein the track is selected from a plurality of tracks by:
sorting the plurality of tracks based on a quality into a sorted list; and
selecting the track based on the sorted list.
14. The system of claim 10 , wherein the track is selected from a plurality of tracks by:
sorting the plurality of tracks based on a taste profile of a user into a sorted list; and
selecting the track based on the sorted list.
15. The system of claim 10 , wherein the track is selected based on a play list associated with the input text.
16. The system of claim 10 , wherein the track is selected from a plurality of tracks by:
determining a suitability of each track in the plurality of tracks, including by determining how clearly a lyric can be heard; and
selecting the track based on the suitability.
17. The system of claim 10 , wherein the audio snippet contains one or more words of the first text set and, and wherein the combined audio contains the one or more words of the first text set and the at least one word of the second text set.
18. The system of claim 10 , wherein splitting the input text into the first text set and the second text set comprises splitting the input text into the first text set, the second text set, and a third text set, and wherein the text-to-speech engine is further configured to:
extract a second audio snippet from a second track based on the third text set, wherein the combined audio further includes the second audio snippet.
19. The method of claim 1 , wherein the first text set being associated with the suitable lyrics or the suitable lyrical characteristic matched by the key value pair in the dictionary structure comprises the first text set being in a title of the track.
20. The system of claim 10 , wherein the first text set being associated with the suitable lyrics or the suitable lyrical characteristic matched by the key value pair in the dictionary structure comprises the first text set being in a title of the track.