The AI4LAM Speech-to-Text Working Group invites you to its next meeting on Tuesday, October 28 at 09:00 US-Pacific | 12:00 US-Eastern | 16:00 UK | 17:00 Central Europe | 03:00 +1 Canberra.
(Note that this will be the odd week of the year when the US is one hour less behind Europe than usual. Please look closely at the time!)
Our topic: ASR for languages other than English
Because many of our group members speak (only) English and work with collections primarily in English, we tend to pay a lot of attention to English language speech-to-text. However, ASR systems that work well for one language do not always work well for another.
So, this week we'll focus on ASR for languages other than English. Three people have graciously agreed to share their experiences with us:
-
Christy Bailey-Tomecek (Beinecke Library, Yale University, USA) on Sonix and aTrain for French, German, Russian, and Hebrew
-
William Mattingly (Yale University, USA) on finetuning with Bantu languages and transfer learning from Swahili
-
Andrija Sagic (Milutin Bojić Library, Serbia) on transcribing Serbian with both Cyrillic and Latin writing systems
In advance of the meeting (whether or not you can attend), please consider filling out this short, anonymous survey about what languages you are trying to transcribe:
https://forms.gle/VxTxeK1oSk4zRepL9
If we get enough responses, we can discuss them during the meeting.
Please join us!
Cheers,
Owen
(on behalf of the Speech-to-Text WG conveners)
|
Owen King
(he/him)
|
|
Metadata Operations Manager
|
|
E:
owen...@wgbh.org
|
|
One Guest Street, Boston,
MA
02135
|
|
|