Method and Apparatus to Model and Transfer the Prosody of Tags across Languages

• US 20140200892A1
• Filed: 01/17/2013
• Published: 07/17/2014
• Est. Priority Date: 01/17/2013
• Status: Active Grant

• Current Assignee: Speech Morphing Systems, Inc.
• Original Assignee: Speech Morphing Systems, Inc.
• Inventors: Yassa, Fathy, Henton, Caroline
• Granted Patent: US 9,418,655 B2
• Time in Patent Office: Days
• US Class Current: 704/257
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/10 Prosody rules derived from text; Stress or intonation
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 2025/906 Pitch tracking

Abstract

Identify, Capture, Retain and Synthesize Non-Linguistic and Discourse Components of Speech across Languages

Claims

1. A method and Apparatus to Model and Transfer the Prosody of Tags across Languages comprising the steps of a first person in speaking in language one (L1);

where the L1 speech is recognized by the ASR;

searching the speech for a known tag;

searching the pieces of text that have common, cpnsistent, or idiomatic intonation patterns, translating the text to language number two (L2);

examine the speech signal of L! to find the segments that correspond to the tag;

extract the fundamental frequency from those sigments and fit a smooth contour such as a cubic spline;

map the stylized smooth contour into the corresponding part of the pitch range of the intended L2 synthesized speech;

stretch or contract stylized smooth contour over time because the duration of the translation will be different;

align the contour with the corresponding L2 segments and impose it on the synthesized L2 speech.