
NO.:
W62410 API
VERSION:
A
PAGE
6
The W62410 allows independent control of the
sensitivity of the VAD, DTMF and Call Progress
Tone detectors.
Self Test and Initialization:
During startup the W62410 checks each time the
state of the storage memory and checks if the
storage memory is corrupt or not (e.g. failure of
the battery backup for DRAM during a power
failure). If the storage memory has become
corrupt then the CHIPSET will automatically
reinitialize the storage memory and its contents
will be lost.
Each time the CHIPSET is reset after a power
down mode the CHIPSET checks if there was a
battery backup failure. In case of Flash memory,
the messages will be preserved since Flash
memory is non volatile. In case of DRAM
memory, the DRAM will be reinitialized.
The CHIPSET supports a function to simulate a
power up (i.e. a cold reset) that forces a complete
re-initialization of the external memory (even
for Flash memory).
Voice Prompt generation:
The W62410 implements high quality voice
prompt playback using the built-in speech
decoder. The CHIPSET has on board 32 Kbytes
for voice prompt storage (which translates to
about 54 seconds of high quality of speech).
To let the sound of the concatenated voice
prompt words sound more fluent, the host can
program a varying length of pause between each
word. This pause can be different between each
word. The result is a less robotic (synthetic) and
therefore a much more naturally sounding voice
prompt message.
If the application requires this, an external voice
prompt ROM can be connected to the CHIPSET,
disabling the internal voice prompt ROM.
When if 54 seconds of voice prompt playback is
not enough, the CHIPSET has built-in support to
drive up to two external voice prompt ROM’s
together with the built-in voice prompt ROM.
For this, two IO pins of the 16 available IO pins
are used (in total 13 IO pins of the available 16
IO pins are freely available to the user). This
gives a total of nearly 3 minutes of voice prompt
speech.
The number of voice prompts supported per 32
Kbytes ROM is set to 512. The theoretical limit
The above information is the exclusive intellectual property of Winbond Electronics and shall not be
disclosed, or reproduced without permission from Winbond.
on the number of voice prompts with one
internal ROM and two external ROM’s is thus
1536.
If wished the user can also put his own choice of
voice prompts into the internal voice prompt
ROM by a simple remasking of the voice prompt
ROM part of the CHIPSET. This can save, for
large quantities, the cost of having to use an
external voice prompt ROM.
Reporting to the host:
The W62410 is capable of reporting the
following statuses back to the host (remember
that, due to the master – slave concept, the host
always has to take the initiative before getting a
status back):
1. The total available remaining recording time
presented in hours, minutes and seconds. The
returned time is a worst case estimate as if
the remaining recording space would be
recorded without silence management at a
fixed bitrate of 4800 bps. Typically, when
silence management is switched on the
recording time will increase between 20%
and 30%.
2. For each mailbox, it can return the number of
present ICM and memo messages.
3. It can return an ID number of the message the
current message pointer is pointing to. This
can be useful information to display to the
user. The CHIPSET automatically takes into
account which mailbox is currently in use to
return the correct ID number.
4. During recording or playback of any
message, it can return a progress indicator.
The number returned is the number of 30
milliseconds frames that have passed since
the start of the recording or playback of the
message. This information can be useful e.g.
for indicating to the user how many seconds
already have been passed while he’s listening
to a message.
5. At any moment, it can return information
about the status of the telephone line.
6. Since the CHIPSET has a built-in Real Time
Clock, the host can request at any moment
the current time.
Near-End Echo Cancellation:
A near-end echo canceller is activated during
playback, tone generation and voice prompt