Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
adrianbg
on Oct 25, 2017
|
parent
|
context
|
favorite
| on:
Speech Recognition Is Not Solved
Yes, exactly. I do wonder whether a similarly good end-to-end system could be trained by constraining the alignments as I've seen done in some papers.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: