Method of upgrading a data stream of multimedia data

• US 7,092,873 B2
• Filed: 01/07/2002
• Issued: 08/15/2006
• Est. Priority Date: 01/09/2001
• Status: Expired due to Fees

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Mengel, Andre, Mecke, Sven, Engelsberg, Andreas, Wollborn, Michael, Kussmann, Holger
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Vo, Huyen X.
• Application Number:
  US10/040,648
• Publication Number:
  US 20020128813A1
• Time in Patent Office: 1,681 Days
• Field of Search: 704/260, 704/258, 704/262, 704/259, 704/261, 704/266, 704/270, 704/200.1, 345/619, 379/88.16, 379/88.14
• US Class Current: 704/200.1
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

Abstract

For upgrading a data stream of multimedia data, which comprises features with textual description, a set of phonetic translation hints is included in the data stream, which specifies the phonetic transcription of parts or words of the textual description. The phonetic transcriptions need not be repeated for each occurrence of a word. This reduces the amount of data necessary for storing or transmitting the description text.

Claims

1. A method of upgrading a data stream of multimedia data, said data stream comprising features with a textual description, said textual description comprising a plurality of words, said method comprising the steps of:

a) including a set of phonetic translation hints in the data stream of the multimedia data in addition to the textual description, wherein each of said phonetic translation hints comprises a repeated word of the textual description and a phonetic transcription of said repeated word, and each of said phonetic translation hints is provided only once in said data stream, wherein said phonetic transcription of said repeated word determines pronunciation of said repeated word and is valid for said textual description without requiring repetition of said phonetic transcription hint for said repeated word at each occurrence of said repeated word in said textual description; and

b) using each of said phonetic transcription hints provided in the data stream to define pronunciation of said repeated word associated therewith at each occurrence of said repeated word in said textual description.

2. The method according to claim 1, wherein said phonetic translation hints are embedded in an MPEG data stream associated with textual type descriptors.

3. The method according to claim 2, whereIn said MPEG data stream is an MPEG-7 data stream.

4. The method according to claim 1, further comprising referring to an alphabet in a predetermined format for representation of phonetic transcription information.

5. The method according to claim 4, wherein said alphabet is an international phonetic alphabet or SAMPA.

6. The method according to claim 1, wherein said phonetic translation hints include a limited number of phonemes.

7. The method according to claim 6, wherein said phonemes are represented with a binary fixed length or variable length code.

8. The method according to claim 7, wherein coding of said phonemes takes into account statistics of the phonemes.

9. The method according to claim 1, further comprising storing said phonetic translation hints in a speech recognition system to better identify corresponding elements of the textual description.

10. The method according to claim 9 wherein the phonetic translation hints together with the corresponding elements of the textual description are implemented in text-to-speech interfaces, speech recognition devices, navigation systems, audio broadcast equipment or telephone applications, in which said textual description is used In combination with phonetic information for search or filtering of information.

11. A method of upgrading a data stream of multimedia data, said data stream comprising features with a textual description, said textual description comprising a plurality of words including a repeated word, said method comprising the steps of:

a) specifying at least a part of said textual description in which said repeated word is repeated at least once;

b) providing a phonetic translation hint for said repeated word only once in said data stream, wherein said phonetic translation hint comprises said repeated word and a phonetic transcription of said repeated word, said phonetic transcription defining pronunciation of said repeated word, so that said phonetic translation hint is not repeated in said at least a part of said textual description at each occurrence of said repeated word, for which the phonetic transcription is given; and

c) using said phonetic transcription provided in said phonetic translation hint to define pronunciation of said repeated word at each occurrence of said repeated word in said at least a part of said textual description.