[Eyes-Free Shell] Remove Pre-recorded Audio

[GoogleCodeExporter]

The shell's use of two voices brings along with it a number of usability
and aesthetic issues. Synthesized and sampled speech talking at two
separate rates, volumes and pitches makes for an inconsistent experience
wherein some parts of speech are too loud while others are too quiet. I
tend to have my phone's TTS volume set low due to privacy concerns and
respect for those nearby, but the sampled speech is too quiet for this and
I end up missing half of my phone's data points. It's also quite noticable
in the Youtube videos wherein the synthesized speech clips quite a bit
because the phone's volume must be cranked to pick up the quiet samples.

I've patched my shell to not use any sampled speech except for the
backspace, because I'm not instantly sure how to set pronunciation for a
given string. The experience is much better, faster and more consistent.

Original issue reported on code.google.com by no...@thewordnerd.info on 11 Nov 2009 at 5:43

Attachments:

• onevoice.patch

[GoogleCodeExporter]

This really needs to be fixed for the version we download from the market. Why 
was 
sampled speech used in the first place? When people get used to the synthesizer 
they 
have chosen it is much easier to understand everything. Adding in a completely 
different voice that sounds more like an echoing robot and uses a lower sound 
volume 
just screams that someone was seriously NOT thinking.
Please fix this!

Original comment by bet...@gmail.com on 22 May 2010 at 3:45

[GoogleCodeExporter]

bethko,
Thank you for reporting this issue to us.  We have made the appropriate fixes 
which 
will be included across all eyes-free project applications during our next 
Market and 
open source push cycle.

Thanks,
Casey Burkhardt

Original comment by caseybur...@google.com on 5 Jun 2010 at 1:46

• Changed title: [Eyes-Free Shell] Remove Pre-recorded Audio
• Changed state: Fixed
• Added labels: Usability