Sign in

username:

password:



Not a member?

Search speechcoding



Search tips

Subscribe to speechcoding



speechcoding by Keywords

ACELP | ADPCM | AMBE | AMR | AMR-NB | CELP | Codebook | DTMF | G.723 | G.726 | G.729 | GSM | Interpolation | LPC | LSF | LSP | MELP | PCM | Perceptual | Pitch | PSOL | QCELP | Quantization | SMV | VAD | Vocoder

Ads

Discussion Groups

Discussion Groups | Speech Coding | Re: Aysemetric Speech Encoding

Technical discussions related to Speech Coding (all itu and other vocoders, ACELP, CELP, AMR, etc)

  

Post a new Thread

Aysemetric Speech Encoding - seandunlevy - Sep 13 16:30:28 2008



I'm looking into a low bit-rate speech compression route that I can 
decompress using an ARM7 TDMI running at 33MHz. There will be no real-
time encoding required, so coding time (as long as it's not months) is 
not important.
Virtually ALL encoders are designed around bidirectional sound whereas 
this is purely for playback.

If anyone has any clues, I would be most grateful.

Oh, I'm a low, low ever so low level coder so all the ARM stuff will be 
written in assembly language and tweaked for cache efficiency.
------------------------------------



(You need to be a member of speechcoding -- send a blank email to speechcoding-subscribe@yahoogroups.com )

Re: Aysemetric Speech Encoding - MAK - Sep 14 8:58:50 2008

Hi, seandunlevy.
> I'm looking into a low bit-rate speech compression route that I can 
> decompress using an ARM7 TDMI running at 33MHz. There will be no real-
> time encoding required, so coding time (as long as it's not months) is
> not important.
> Virtually ALL encoders are designed around bidirectional sound whereas
> this is purely for playback.

The simplest solution for you is ADPCM (can be as low as 2-bit per sample)
To get better compression you can use time scale modification like TDHS.
I'm not sure if the 33MHz ARM can run recent codecs with performance requirements.
Also most recent codecs use analysis-by-sinthesis method,
so the encoder and decoder performances are quite the same.
MAK                          mailto:m...@mail.ru
------------------------------------



(You need to be a member of speechcoding -- send a blank email to speechcoding-subscribe@yahoogroups.com )