Neural text-to-speech synthesis with multi-level text information
AssigneeMicrosoft Technology Licensing, LLC
InventorsHuaiping MING, Lei HE
Life of this patent
Estimating the term…
Abstract
A method and apparatus for generating speech through neural text-to-speech (TTS) synthesis. A text input may be obtained (1310). Phoneme or character level text information may be generated based on the text input (1320). Context-sensitive text information may be generated based on the text input (1330). A text feature may be generated based on the phoneme or character level text information and the context-sensitive text information (1340). A speech waveform corresponding to the text input may be generated based at least on the text feature (1350).
Claims20
A method for generating speech through neural text-to-speech (TTS) synthesis, comprising:
- obtaining a text input;
- generating phoneme or character level text information based on the text input;
- generating context-sensitive text information based on the text input;
- generating a text feature based on the phoneme or character level text information and the context-sensitive text information;
- and
- generating a speech waveform corresponding to the text input based at least on the text feature.
An apparatus for generating speech through neural text-to-speech (TTS) synthesis, comprising:
- a text input obtaining module, for obtaining a text input;
- a phoneme or character level text information generating module, for generating phoneme or character level text information based on the text input;
- a context-sensitive text information generating module, for generating context-sensitive text information based on the text input;
- a text feature generating module, for generating a text feature based on the phoneme or character level text information and the context-sensitive text information;
- and
- a speech waveform generating module, for generating a speech waveform corresponding to the text input based at least on the text feature.
A neural text-to-speech (TTS) system, comprising:
- a multi-input encoder, configured for generating a text feature corresponding to a text input;
- a decoder, configured for generating acoustic features corresponding to the text feature through an attention mechanism;
- and
- a vocoder, configured for generating a speech waveform corresponding to the text input based on the acoustic features,
- wherein the multi-input encoder is further configured for generating the text feature through:
- generating phoneme or character level text information based on the text input,
- obtaining context-sensitive text information corresponding to the text input, and
- generating the text feature based on the phoneme or character level text information and the context-sensitive text information.
An apparatus for generating speech through neural text-to-speech (TTS) synthesis, comprising:
- at least one processor;
- and
- a memory storing computer-executable instructions that, when executed, cause the at least one processor to:
- obtain a text input,
- generate phoneme or character level text information based on the text input,
- generate context-sensitive text information based on the text input,
- generate a text feature based on the phoneme or character level text information and the context-sensitive text information, and
- generate a speech waveform corresponding to the text input based at least on the text feature.
Dependencies are inferred from the claim text.