System and method for generating customized text-to-speech voices
View Patent ↗A system and method are disclosed for generating customized text-to-speech voices for a particular application. The method comprises generating a custom text-to-speech voice by selecting a voice for generating a custom text-to-speech voice associated with a domain, collecting text data associated with the domain from a pre-existing text data source and using the collected text data, generating an in-domain inventory of synthesis speech units by selecting speech units appropriate to the domain via a search of a pre-existing inventory of synthesis speech units, or by recording the minimal inventory for a selected level of synthesis quality. The text-to-speech custom voice for the domain is generated utilizing the in-domain inventory of synthesis speech units. Active learning techniques may also be employed to identify problem phrases wherein only a few minutes of recorded data is necessary to deliver a high quality TTS custom voice.
1. A method comprising:
when a user is determined to be new, changing a front-end which converts text into linguistic tokens and tags the linguistic tokens with a prosody, a pronunciation, a speech act, and an emotion of the text, to yield a user-specific front-end;
in response to a user request from the user to generate a custom text-to-speech voice for a vehicle, the custom text-to-speech voice being associated with a domain, automatically:
collecting, via a processor, text data associated with the domain from a pre-existing text data source, to yield collected text data;
selecting, based on the user specific front end, synthesis speech units specific to the domain from a pre-existing inventory of synthesis speech units using the collected text data;
caching, via the processor, the synthesis speech units specific to the domain as an in-domain inventory of synthesis speech units; and
generating, via the processor, the custom text-to-speech voice for a specific task in the domain utilizing the in-domain inventory of synthesis speech units.
2. The method of claim 1 , further comprising determining whether the custom text-to-speech voice conforms to a selected level of synthesis quality.
3. The method of claim 2 , further comprising:
when the custom text-to-speech voice does not conform to the selected level of synthesis quality, collecting additional text data associated with the domain.
4. The method of claim 3 , further comprising iteratively collecting additional text data until the custom text-to-speech voice conforms to the selected level of synthesis quality.
5. The method of claim 1 , wherein the pre-existing text data source is one of a domain-related website, e-mail and transcriptions of conversations.
6. The method of claim 1 , wherein the pre-existing text data source is a sector-related website.
7. The method of claim 6 , further comprising categorizing websites by sector to identify websites as pre-existing text data sources prior to collecting the text data.
8. The method of claim 1 , wherein collecting text data associated with the domain from the pre-existing text data source further comprises mining specific words and phrases from the pre-existing text data source.
9. The method of claim 8 , wherein mining specific words and phrases from the pre-existing text data source further comprises mining specific words and phrases using an n-gram selection.
10. The method of claim 8 , wherein mining specific words and phrases from the pre-existing text data source further comprises mining specific words and phrases using a maximal mutual information approach.
11. The method of claim 1 , further comprising manually adding one of relevant words and relevant phrases to the collected text data for use in generating the in-domain inventory of synthesis speech units.
12. The method of claim 1 , further comprising applying active learning to identify one of problematic speech units and problematic phrases within the in-domain inventory of synthesis speech units.
13. The method of claim 12 , further comprising:
recording one of words and phrases according to the one of problematic speech units and problematic phrases; and
integrating the one of words and phrases into the in-domain inventory of synthesis speech units.
14. The method of claim 13 , further comprising:
determining whether the custom text-to-speech voice conforms to a selected synthesis quality.
15. The method of claim 14 , further comprising:
when the custom text-to-speech voice does not conform to the selected level of synthesis quality, collecting additional text data associated with the domain.
16. The method of claim 14 , wherein when the custom text-to-speech voice does not conform to the selected synthesis quality, recording one of additional words and additional phrases to increase the in-domain inventory of synthesis speech units.
17. The method of claim 12 , further comprising:
determining a minimal in-domain inventory for recording to meet a selected custom voice synthesis quality;
based on the minimal in-domain inventory, recording one of words and phrases according to the one of problematic speech units and problematic phrases; and
integrating the one of words and phrases into the in-domain task-independent inventory of synthesis speech units.
18. The method of claim 13 , further comprising:
determining whether the custom text-to-speech voice conforms to a custom voice synthesis quality; and
when the custom text-to-speech voice does not conform to the custom voice synthesis quality, recording one of additional words and additional phrases to increase the in-domain inventory of speech units.
19. 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:
when a user is determined to be new, changing a front-end which converts text into linguistic tokens and tags the linguistic tokens with a prosody, a pronunciation, a speech act, and an emotion of the text, to yield a user-specific front-end;
in response to a user request from the user to generate a custom text-to-speech voice for a vehicle, the custom text-to-speech voice being associated with a domain, automatically:
collecting, via a processor, text data associated with the domain from a pre-existing text data source, to yield collected text data;
selecting, based on the user specific front end, synthesis speech units specific to the domain from a pre-existing inventory of synthesis speech units using the collected text data;
caching, via the processor, the synthesis speech units specific to the domain as an in-domain inventory of synthesis speech units; and
generating, via the processor, the custom text-to-speech voice for a specific task in the domain utilizing the in-domain inventory of synthesis speech units.
20. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
when a user is determined to be new, changing a front-end which converts text into linguistic tokens and tags the linguistic tokens with a prosody, a pronunciation, a speech act, and an emotion of the text, to yield a user-specific front-end;
in response to a user request from the user to generate a custom text-to-speech voice for a vehicle, the custom text-to-speech voice being associated with a domain, automatically:
collecting, via the computing device, text data associated with the domain from a pre-existing text data source, to yield collected text data;
selecting, based on the user specific front end, synthesis speech units specific to the domain from a pre-existing inventory of synthesis speech units using the collected text data;
caching, via the computing device, the synthesis speech units specific to the domain as an in-domain inventory of synthesis speech units; and
generating, via the computing device, the custom text-to-speech voice for a specific task in the domain utilizing the in-domain inventory of synthesis speech units.