<feed xmlns='http://www.w3.org/2005/Atom'>
<title>TaSTT.git/Scripts/requirements_frozen.txt, branch v0.19.0</title>
<subtitle>Free self-hosted STT for VRChat.</subtitle>
<id>https://git.yummers.dev/TaSTT.git/atom?h=v0.19.0</id>
<link rel='self' href='https://git.yummers.dev/TaSTT.git/atom?h=v0.19.0'/>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/'/>
<updated>2024-06-06T01:15:47+00:00</updated>
<entry>
<title>Upgrade faster-whisper with flash-attention2</title>
<updated>2024-06-06T01:15:47+00:00</updated>
<author>
<name>yum</name>
<email>yum.food.vr@gmail.com</email>
</author>
<published>2024-06-06T01:15:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/commit/?id=4f0fb5b17de990517e3c1de7ffee5d0f3c9a8961'/>
<id>urn:sha1:4f0fb5b17de990517e3c1de7ffee5d0f3c9a8961</id>
<content type='text'>
This should be significantly more efficient than prior versions.

* add large-v3 &amp; distilled variant
* simplify model acquisition code now that distilled models are part of
  faster-whisper.
</content>
</entry>
<entry>
<title>Pin huggingface_hub to 0.16.4</title>
<updated>2023-09-11T20:53:14+00:00</updated>
<author>
<name>yum</name>
<email>yum.food.vr@gmail.com</email>
</author>
<published>2023-09-11T20:46:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.yummers.dev/TaSTT.git/commit/?id=0447f37fb744a1b350f6b92e4d140dbdb1c8d3ec'/>
<id>urn:sha1:0447f37fb744a1b350f6b92e4d140dbdb1c8d3ec</id>
<content type='text'>
0.17.x are breaking faster_whisper's ability to download models.

Also:
* Start using frozen requirements.txt.
* Conditionally install torch &amp; legacy whisper only when doing
  mechanical optimization.
</content>
</entry>
</feed>
