Abstract: A speech-processing system receives input data representing text. An encoder processes segments of the text to determine embedding data, and a decoder processes the embedding data to determine one or more categories associated with each segment. Output data is determined by selecting words based on the segments and categories.

Abstract: A system may determine text for inclusion in a voice corpus for use in text-to-speech (TTS) processing using an interface that allows multiple entities to connect and review potential text segments concurrently. The interface may allow networked communication with the system. Individual text segments may be approved or rejected by reviewing entities, such as proofreaders. The system may prioritize text segments to send to reviewers and may re-prioritize text segments based on a linguistic coverage of previously accepted text segments.

Abstract: A text-to-speech (TTS) system may be configured to incorporate breath sounds in the output speech. By incorporating breath sounds into speech output from text a TTS system may be able to mimic more naturally sounding human speech, particularly for long-form narration of text longer than short phrases. The breath sounds may be stored as units for unit selection or may be generated during parametric synthesis. The acoustic features of the breath sounds and duration between breaths may depend upon the punctuation of text, the linguistic distance between breaths, the breaks between intonational phrases, the linguistic context of the breaths, and other factors.