Kevin Chiu

There’s a new startup called Jott.

Basically, you call a number, they transcribe what you say and then send it to you in an email. Unfortunately, the messages you leave are transcribed by humans! (privacy, scale issues)

Here’s what Jott’s CEO has to say about it:

jotts are completely anonymized to transcribers. They do not know your account name or other information.

Thanks,
John
CEO, Jott.com

Perhaps they should just reroute calls through Vista’s voice recognition. It’s probably the first widely available voice recognition implementation that’s up to the task.

Update: Please use this tutorial instead. It’s a bit lengthier, but it works with the current version of vista. (10/3/07)

I’ve found an easier way to do the Volume Mixer trick in Windows Vista. Just open up the volume mixer and select the output you want to use from the device menu. Then launch the applications you want to assign to the currently viewed audio device. Repeat for each audio device you want to use.

In this tutorial, we present a method for individually assigning audio sources to different audio outputs.

1. Right click the sound icon in the taskbar and select Playback Devices.

2. Make the destination audio output the default device.

3. Launch the audio sources you wish to assign to the output you just made default.

Repeat steps 2 and 3 for each audio output you want to use.

Open up the volume controls once you’re finished tinkering, and you should end up with something like this:

Headphone-bound applications

Speaker-bound applications

The trick: Each time you open an application, its sound will be routed to whatever output you have currently selected as the default. This pairing between source and output is “sticky.” Changing the default output does not migrate any of the sources to the new default output.

Microsoft should make this easier. A 2D grid of audio inputs/outputs would be optimal, but perhaps the easiest thing to implement at this stage is a “move to” contextual menu item, which moves the selected audio source to a specified audio output.