Great find Pam..A key feature of portable MP3 players is the increasing memory that allows hundreds of musical tracks to be stored and played back.
Manually navigating through such a large number of audio selections using a small liquid crystal display (LCD) will prove tedious and
unsatisfactory to most users. With Bell Labs' voice recognition software, users of portable MP3 players would have the option of
selecting their favorite artist and musical track from an intuitive, highly accurate voice driven menu. The voice driven menu would be
setup with jukebox or audio manager PC software allowing users to efficiently customize keywords for main menu headings and
subheadings according to individual preferences.
During the setup of the voice driven menu, the jukebox or audio manager software can exploit the ID3 tag information to make the
process of entering artist and track names a simple "point and click" process for the consumer. In addition, as the various information is
entered into the voice driven menu by the user, the jukebox or audio manager software can automatically run the linguistic module of
Bell Labs' distributed Text-to-Speech (TTS) software to convert the information into speech prompts as described further below. Of
course, Bell Labs' voice recognition software can also be used to control other player functions, such as tone or equalization settings,
pause, replay, skip a track, etc., if so desired.
A significant feature of Bell Labs' distributed TTS software is that it enables appropriately processed ID3 tag (or, optionally, user
supplied) information to be converted into speech, thus giving the MP3 player a "voice" that can prompt the user by naming artists or
song titles from various categories.
This would work as follows:
The standard ID3 tag (or optional user supplied) information entered into the voice driven menu is processed by the linguistic
processing module of Bell Labs' TTS software which is an application that runs under the jukebox or audio manager PC software. This
converts the ID3 tag information into compact phonetic information-a "phonetic ID3 tag"-that is included with the standard ID3 tag
information of each MP3 file. After the MP3 files and tag information are transferred to the player, the synthesis portion of the TTS
software runs on the player's DSP and converts the ID3 phonetic information of the particular track into speech that the user would
hear over the headphones.
Voice Interaction Samples
Sample Dialog 1:
MP3 Player: "Name the artist or band or music type"
User: "Sting"
MP3 Player: "CD or song?"
User: "Song"
MP3 Player: "Which song?"
User: "Every Breath You Take"
[MP3 Player starts playing a song "Every Breath You Take" by Sting]
Sample Dialog 2:
MP3 Player: "Name the artist or band or music type"
User: "Celine Dion"
MP3 Player: "CD or song?"
User: "CD"
MP3 Player: "Which CD?"
User: "What are my choices?"
MP3 Player: "Your choices are: "Falling Into You", "Let's Talk About Love", "The Color of My Love" …
User: "Let's Talk About Love"
[MP3 Player starts playing a track from "Let's Talk About Love" CD by Celine Dion]
User: "Next Song"
[MP3 Player starts playing the next song from the CD]
Contact Information For more information, contact the Lucent Technologies Intellectual Property Business/Software Solutions Group
at 1-800-462-8146 or 1-610-992-1577 or send e-mail to mobilemedia@lucent.com, or visit www.lucentssg.com.
AI
