System and method for blending synthetic voices
AssigneeAT&T Intellectual Property II, L.P.
InventorsDavid A. Kapilow, Kenneth H. Rosen, Juergen Schroeter
Life of this patent
Estimating the term…
Abstract
A system and method for generating a synthetic text-to-speech TTS voice are disclosed. A user is presented with at least one TTS voice and at least one voice characteristic. A new synthetic TTS voice is generated by blending a plurality of existing TTS voices according to the selected voice characteristics. The blending of voices involves interpolating segmented parameters of each TTS voice. Segmented parameters may be, for example, prosodic characteristics of the speech such as pitch, volume, phone durations, accents, stress, mis-pronunciations and emotion.
Claims34
A method of generating a synthetic voice comprising:
- receiving a user selection of a first text-to-speech (TTS) voice and a second TTS voice from a plurality of TTS voices;
- receiving at least one user-selected voice characteristic;
- and
- generating a new TTS voice by blending the first TTS voice and the second TTS voice and according to the at least one user-selected voice characteristic.
A method of generating a synthetic voice, the method comprising:
- receiving a user selection of a TTS voice and a voice characteristic;
- and
- presenting the user with a new TTS voice comprising the selected TTS voice blended with at least one other TTS voice to achieve the selected voice characteristics.
A system for generating a synthetic voice, the system comprising:
- a module for presenting a user with a plurality of TTS voices to select at least one voice characteristic;
- a module for receiving a user-selected first TTS voice, a user-selected second TTS voice, and at least one user-selected voice characteristic;
- and
- a module for generating a new TTS voice by blending the first TTS voice and the second TTS voice and according to the at least one user-selected voice characteristic.
A method of generating a text-to-speech (TTS) voice generated by blending at least two TTS voices, the method comprising:
- establishing a voice profile for each of a plurality of TTS voices, each voice profile having speaker-specific parameters;
- receiving a request for a new TTS voice from a user;
- and
- generating the new TTS voice by blending speaker-specific parameters obtained from the voice profiles for at least two TTS voices.
A test-to-speech (TTS) voice generated from a method of blending at least two TTS voices, the method comprising:
- establishing a voice profile for each of a plurality of TTS voices, each voice profile having speaker-specific parameters;
- receiving a request for a blended TTS voice from a user;
- and
- generating the blended TTS voice by blending speaker-specific parameters obtained from the voice profiles for at least two TTS voices.
Dependencies are inferred from the claim text.