(This document is the original specification for the creation of Audrey's sound scheme. The sound scheme was then created in accordance with it.)

Audrey's Language

Since the sounds that Audrey makes will be like a spoken language, the language should be as simple as possible but without oversimplification. This makes the language easy to understand and therefore reduces time and effort needed to learn Audrey's language.

Sound Categories

Start Up / Wake Up

- Possible random sequence of entire vocabulary of system sounds
- Possible different start up / wake up sound sequence for day and night
- User can set day/night transition time.
The start up, wake up, and sleep sounds are key in making Audrey have personality since it is how she will greet the user. This is also the opportunity for a branding moment, though the branding should probably be limited to a very brief moment so the flow of daily use is not hindered by it. This moment could be sound or a combination of sound and visual.

Selecting with buttons or stylus

- Maintain consistency between methods of selection
There should be a general selection sound since the idea is the same whenever selecting, whether using buttons or the stylus. The main applications could also have a sound associated with them to reinforce their selection and their central role.

Selecting menus

- Provide feedback for opening and collapsing of menus
Since the menus have an extra actions associated with them, beyond a simple click, the menus should have an opening and collapsing sound to reinforce that action.

Selecting channels via dial

- A subtle sound that strengthens the uniqueness of the dial navigation
- Could be related to the sound of tuning in a radio frequency 
        as if it passing through additional content between channels.
Since the channels are distinctly separate part of Audrey, selecting channels using the channel knob should have a distinct sound associated with it.

Data Entry

- Different sounds for keyboard and stylus
Data Entry should have a very subtle sound to reinforce the actions, like the sound a keyboard or pen makes when using it. There should be different sounds for stylus and keyboard since the action is very different.

Editing (Cut / Copy / Paste)

- Cut = sound A
- Copy = sound A+B
- Paste = sound B
Editing functions should have sound feedback to reinforce whether the action is completed, especially with a Copy since it doesn't change anything visually. Editing should have two sounds, an Cut sound and a Paste sound. The Copy sound would be a combination of the two sounds since a Copy is like a combined Cut and Paste.

Toss/Delete Sound

- reinforcing sound for any delete/toss action
Since deleting is a unique action that can't be undone, the deletion action should be reinforced by a unique sound.

Processing / Waiting / Loading

- Consistent and supportive sound that is used for web connection, hot sync, 
        or file loading delays.
Audrey should make a sound that communicates that she is working on the task at hand, whether its connecting to the internet, waiting for a buy transaction to finish, hot-syncing a Palm, installing a new program, etc. This will give the user feedback as to the state of the current request without the user having to walk up to the device to look at the screen. This sound is needed in addition to the completion sound because it allows the user to walk away and come back and know that Audrey is still working or not without wondering whether they might have missed the completion sound.

Completion of task

- A consistent sound to signal completion of tasks such as transfer of mail, 
        transfer of Palm data, internal saving of data.
Audrey should make a completion sound to communicate to the user that their request has been fulfilled. At this point the user might want to go back to Audrey and start using it again.


- Falling into several categories, guides present various text messages including:

1) confirmations - uh huh, wow, cool
2) hellos, goodbyes - good morning, how are you?, bye, ciao 
3) alerts - pardon me, oops, 
4) reminders - have you checked?, did you notice?
5) referrals - try this, did you notice?, I thought this might interest you
Each of these groups could have a sound that embodies an attitude that is appropriate to the category of intervention. The sounds accompanying the guides should be the most evocative of human language since Audrey is directly trying to tell the user something, rather than providing feedback.

Initial Setup

The initial setup phase should use all of the above sounds heavily. The user will just be getting to know Audrey so she should try to teach the user her language by using the language emphatically and in a clearly distinct manner. Since the setup phase will not be frequently visited, heavy use of sound will not be annoying. The initial setup phase is also an opportunity to try to establish Audrey's vocal persona, through the use of extended sound pieces and maybe also music.

$Id: language.html,v 1.11 2005/08/24 21:09:31 hans Exp $