(Page 3 of 8 in this chapter)

6.2 Voice Encoding Formats

When recording or playing voice files, the application must select an encoding format. The primary issue to consider when selecting a format is the compression ratio versus the fidelity. More aggressive compression requires less disk space and reduces host-to-AG loading.
AG boards support the following encoding formats: NMS ADPCM G.726 compliant ADPCM OKI ADPCM (6 and 8 kHz) mu-law and A-law 8-bit and 16-bit PCM at 11 kHz (mono) 16-bit PCM at 8 kHz
Each encoding has a minimum data block size, called a frame. Frames vary in size and duration depending upon the encoding format. For Natural MicroSystems' boards, a frame corresponds to 10 or 20 milliseconds of speech, depending on the encoding.
The encodings refer to the data going to and from the host, typically stored in a voice file. This host encoding is independent of the line encoding, which is always either mu-law or a-law, depending on how the board is configured when it is initialized. The exception to this is ADI_ENCODE_NMS_64, in which the encoding is derived from the line encoding.
The encoding formats are: Encoding Format Description Sample Size (bits) Sample Rate (Hz) Frame Size (Bytes) Frame Time (ms) Data Rate (Bytes/sec) ADI_ENCODE_NMS_16 NMS ADPCM 16 kbit/s 2 8000 42 20 2100 ADI_ENCODE_NMS_24 NMS ADPCM 24 kbit/s 3 8000 62 20 3100 ADI_ENCODE_NMS_32 NMS ADPCM 32 kbit/s 4 8000 82 20 4100 ADI_ENCODE_NMS_64 framed PCM 64 kbit/s 8 8000 162 20 8100 ADI_ENCODE_MULAW mu-law 64 kbit/s 8 8000 80 10 8000 ADI_ENCODE_ALAW A-law 64 kbit/s 8 8000 80 10 8000 ADI_ENCODE_PCM8M16 PCM 8k mono 16-bit (WAVE format) 16 8000 160 10 16000 ADI_ENCODE_OKI_24 OKI ADPCM 24 kbit/s 4 6000 30 10 3000 ADI_ENCODE_OKI_32 OKI ADPCM 32 kbit/s 4 8000 40 10 4000 ADI_ENCODE_PCM11M8 PCM 11kHz 8 bit mono (WAVE format) 8 11000 110 10 11000 ADI_ENCODE_PCM11M16 PCM 11 kHz 16- bit mono (WAVE format) 16 11000 220 10 22000 ADI_ENCODE_G726 G.726 ADPCM 32 kbit/s 4 8000 40 10 4000

Encoding Format	Description	Sample Size (bits)	Sample Rate (Hz)	Frame Size (Bytes)	Frame Time (ms)	Data Rate (Bytes/sec)
`ADI_ENCODE_NMS_16`	NMS ADPCM 16 kbit/s	2	8000	42	20	2100
`ADI_ENCODE_NMS_24`	NMS ADPCM 24 kbit/s	3	8000	62	20	3100
`ADI_ENCODE_NMS_32`	NMS ADPCM 32 kbit/s	4	8000	82	20	4100
`ADI_ENCODE_NMS_64`	framed PCM 64 kbit/s	8	8000	162	20	8100
`ADI_ENCODE_MULAW`	mu-law 64 kbit/s	8	8000	80	10	8000
`ADI_ENCODE_ALAW`	A-law 64 kbit/s	8	8000	80	10	8000
`ADI_ENCODE_PCM8M16`	PCM 8k mono 16-bit (WAVE format)	16	8000	160	10	16000
`ADI_ENCODE_OKI_24`	OKI ADPCM 24 kbit/s	4	6000	30	10	3000
`ADI_ENCODE_OKI_32`	OKI ADPCM 32 kbit/s	4	8000	40	10	4000
`ADI_ENCODE_PCM11M8`	PCM 11kHz 8 bit mono (WAVE format)	8	11000	110	10	11000
`ADI_ENCODE_PCM11M16`	PCM 11 kHz 16- bit mono (WAVE format)	16	11000	220	10	22000
`ADI_ENCODE_G726`	G.726 ADPCM 32 kbit/s	4	8000	40	10	4000

6.2.1 Buffer Sizes

All buffers submitted to AG Access for play operations must be able to contain an integral number of frames for the selected encoding format. For example, if you select ADI_ENCODE_NMS_24, the buffer size must be a multiple of 62 bytes. Failure to submit a buffer meeting this size requirement for a play operation causes the function to terminate with a CTAERR_BAD_SIZE reason. Other than the multiple frame size requirement, buffers submitted to AG Access can be any size.
Each AG board has a physical buffer size which is both board- and encoding- dependent. If you submit a buffer larger than the physical size, AG Access divides the buffer into physical segments and submits those segments to the AG board. To eliminate fractional buffers and to reduce the AG-to-host interactions, the optimum user buffer will be a multiple of the physical AG buffer size. This size is retrieved via the function adiGetEncodingInfo.
AG Access employs a double-buffering scheme when recording and playing voice files. This means that when the AG board finishes processing a buffer, the subsequent buffer must already be allocated and submitted to AG Access by the application.

6.2.2 Underruns

Underruns are manifested as gaps in the voice data. On heavily loaded systems, the throughput requirements between the host and the AG board may cause gaps in the voice record or playback. Failure to maintain pace with the AG board can also cause underruns in the record or playback operation. Greater file compression may be necessary to eliminate the problem.
AG Access counts the number of underruns that occur, but not their duration. Retrieve the underrun count by calling adiGetRecordStatus and adiGetPlayStatus.
Note: It is important not to submit small buffers (e.g., buffers that hold less than one second of data) because small buffers can also cause underruns. The data throughput for a given encoding method can be derived from the values returned by adiGetEncodingInfo.

(Page 3 of 8 in this chapter)

tech_support@nmss.com