[vorbis] Vorbis vs Speex

[engdev]

Hi there,

I have also looked very briefly at Vorbis vs. Speex vs. MP3

I am constrained by the application I am using to ACM codecs, of which the
vorbis codec seems broken and furthermore there seems a reluctance to get it
operational, (not surprisingly, the developers have more important things to
do).

So I am left with MP3 and Speex. In actual fact the speex ACM codec fails
when used with some programs (notably trying to use it with CoolEdit, but it
works fine with WinAmp).

MP3 beats Speex at data rates above about 15-20 kbps.
Speex beats MP3 at rates 8kbps to around 15kbps
MP3 won't go as low as speex (all the way to 2.1kbps).

What sort of bitrates were you looking at for your Vorbis/Speex comparisons?

Owen.

----- Original Message ----- 
From: "J C Fitzgerald" <v7022 at wave.co.nz>
To: <vorbis at xiph.org>
Sent: Sunday, September 14, 2003 10:53 AM
Subject: Re: [vorbis] Vorbis vs Speex

<p>> On Mon, May 26, 2003 at 02:23:31PM +1200, J C Fitzgerald wrote:
> >
> > 1.  Which would your experiences show is the better codec for recording
> >     high quality speech such as radio interviews?
> >
> After much experimentation over the last few months, I'm now in a position
> to answer my own question.  I have to say that I agree with "noprivacy"
> when he said:
>
> On Mon, May 26, 2003 at 02:28:06PM -0500, noprivacy at earthlink.net wrote:
> >
> > I've tried it, but I really wasn't impressed with Speex.  mp3 & ogg at
their
> > lowest settings worked better.
>   .....
> > Speex isn't that good.
> >
>
> I found that bit for bit, Vorbis never gives lower quality for speech
> than Speex and it's usually considerably higher.  Vorbis seems more
> suitable for file-based spoken audios than Speex.
>
> There are also a number of non-quality issues to take into account:
>
> *  It is unclear whether Speex will ever be included in hardware devices,
> *  The Speex user base may not have the critical mass needed to ensure its
>    survival,
> *  It seems doubtful that it will ever be widely supported in software,
> *  Unexpected non-spoken components trigger a switch to Vorbis anyway, and
> *  There are advantages to using a single lossy format for all recordings.
>
> I regret these findings because I like the idea of having a format
> specificly targetted at speech, but for my purposes Vorbis seems the
> better choice at present.  I do hope that Speex continues to flourish and
> improve, but suspect that it will do so more in arenas such as telephony.
>
> John.
> --- >8 ----
> List archives:  http://www.xiph.org/archives/
> Ogg project homepage: http://www.xiph.org/ogg/
> To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
> containing only the word 'unsubscribe' in the body.  No subject is needed.
> Unsubscribe messages sent to the list will be ignored/filtered.
>

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.