FS1016 Speech Coder
for the TMS320C5x
by Signals and Software Limited

Software Overview

DSP software which implements US Federal Standard 1016. This codes narrowband speech (300–3400 Hz, 
sampled at 8 kHz) to 4.8 kb/s. By making a proprietary modification to FS1016, a bit rate of 7.2 
kb/s is also provided.

Features and Benefits

The FS1016 specification uses a Codebook Excited Linear-Prediction (CELP) algorithm. The audio 
performance for a single encode/decode at 4.8 kb/s shows up as some degradation from telephone 
quality in terms of harshness and occasional artifacts. The performance is typically better with 
male speakers than with female speakers. The performance at 7.2 kb/s is noticeably improved in that 
less harshness is present. In terms of Mean Opinion Score (MOS) for a single encode/decode, the 
4.8-kb/s rate gives a rating of ~3.1 and the 7.2-kb/s rate a rating of ~3.5.

Processor and System Requirements

FS1016
Program Memory
(Words)
Data Memory
(Words)
Processing Load (MIPS)
Code
Data Tables
Permanent
Temporary*
Encoder + decoder
~4700
~3100
~550
~500
~19
Encoder only
~3500
~3000
~450
~500
~17
Decoder only
~1200
~2600
~80
~500
~2
* Must be in internal RAM
The above figures apply to both the 4.8- and 7.2-kb/s bit rates.

Usage Limitations or Performance Considerations

The code consists of four subroutines: an initialization routine and a processing routine for each 
of the encoder and decoder functions. The initialization routines are normally called only on DSP 
reset. The audio input and output format is 8-kHz linear samples, and the frame size on which the 
processing routines operate depends on the required bit rate. At the 4.8-kb/s rate, the frame size 
is 240 samples (30 ms); at 7.2 kb/s, it is 160 samples (20 ms). Each input frame of speech is 
converted by the encoder into a 144-bit parameter block. The decoder uses this data to reconstruct 
the speech frame. Applications of the algorithm include speech compression for speech storage, 
digital circuit multiplication, telephony multiplexing applications, mixed voice and data 
transmission.

Availability

Now, under licence, for a one-off payment and/or royalties depending on the commercial application. 
An Application Note is available. Support consultancy for code integration is also available.

Company Background and Contact Information

Signals and Software Limited (SASL), based in Harrow, Middlesex, UK is a design consultancy 
specializing in the area of Digital Signal Processing (DSP). From concept and algorithm design 
through to real-time DSP implementation, SASL is able to offer its clients fast and cost-effective 
solutions to their DSP needs. Services include: feasibility studies, DSP software to order, DSP 
research and algorithm design, computer simulations (C or PASCAL), and hardware design.

Contact:	David Morley
		3 Jardine House
		Bessborough Road
		Harrovian Business Village
		Harrow, Middlesex, HA1 3EX
		United Kingdom
		+44 (0) 181 426 9533
		Fax: +44 (0) 181 869 1182
		e-mail: davem@sasl.demon.co.uk