Unlike feedforward neural networks, RNNs can use their internal state (memory) to process sequences of inputs. This makes them applicable to tasks such as unsegmented, connected handwriting recognition or speech recognition.
AKIDA has incremental learning but that is a feature not relevant to processing speech.
AKIDA1000 can pick up individual pre-saved words like "hi", "good" etc with high accuracy but when it comes to a speech containing multiple words, adjectives etc, it will not be able to process it.
Therefore, as I said before, can't be an AI for cloud.
2020 BRN Discussion, page-19977
-
- There are more pages in this discussion • 9,337 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Add BRN (ASX) to my watchlist
 (20min delay)
|
|||||
|
Last
20.0¢ |
Change
-0.005(2.44%) |
Mkt cap ! $371.1M | |||
| Open | High | Low | Value | Volume |
| 20.5¢ | 20.5¢ | 19.5¢ | $843.2K | 4.220M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 5 | 71948 | 20.0¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 20.5¢ | 644737 | 16 |
View Market Depth
| Last trade - 16.10pm 26/07/2024 (20 minute delay) ? |
| BRN (ASX) Chart |