<feed xmlns='http://www.w3.org/2005/Atom'>
<title>TaSTT.git/TaSTT-Whisper, branch v0.9.0</title>
<subtitle>Free self-hosted STT for VRChat.</subtitle>
<id>https://git.yummers.dev/TaSTT.git/atom?h=v0.9.0</id>
<link rel='self' href='https://git.yummers.dev/TaSTT.git/atom?h=v0.9.0'/>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/'/>
<updated>2023-03-06T00:26:18+00:00</updated>
<entry>
<title>Animator generation and dumping mics no longer hang GUI</title>
<updated>2023-03-06T00:26:18+00:00</updated>
<author>
<name>yum</name>
<email>yum.food.vr@gmail.com</email>
</author>
<published>2023-03-06T00:26:18+00:00</published>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/commit/?id=058cd57ed2d18d6e14408881c3229fab7edbadb1'/>
<id>urn:sha1:058cd57ed2d18d6e14408881c3229fab7edbadb1</id>
<content type='text'>
Do these in a std::future.

* SetAffinityMask() now returns a value on all control paths
</content>
</entry>
<entry>
<title>Bugfix: fix use-after-free in GetMicsImpl</title>
<updated>2023-02-28T08:34:10+00:00</updated>
<author>
<name>yum</name>
<email>yum.food.vr@gmail.com</email>
</author>
<published>2023-02-28T08:34:10+00:00</published>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/commit/?id=f65b93aa6f0a5b7ddd069cd1f50f142029e4a5e5'/>
<id>urn:sha1:f65b93aa6f0a5b7ddd069cd1f50f142029e4a5e5</id>
<content type='text'>
* Plumb beam search params into whisper cpp implementation
  (currently broken)
</content>
</entry>
<entry>
<title>Improve behavior around VAD segmentation events</title>
<updated>2023-02-27T04:12:36+00:00</updated>
<author>
<name>yum</name>
<email>yum.food.vr@gmail.com</email>
</author>
<published>2023-02-27T04:12:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/commit/?id=d96851f716b9e18d827ba7795a343dbf3cf529c4'/>
<id>urn:sha1:d96851f716b9e18d827ba7795a343dbf3cf529c4</id>
<content type='text'>
Use forked Whisper implementation which has tweaks to reduce dropped
words around the beginning VAD segments.

* Retain audio after VAD segmentation events
</content>
</entry>
</feed>
