Comments (4)
m-bain
commented on May 22, 2024
Memory scales with length of input segment to wav2vec2 model. With current version, adding
--vad_filter
Ensures that no segments greater than 30s are inputted to the w2v model. You can further reduce gpu reqs by
a. Using smaller w2v align_model
b. Limit input segment duration
I will add b as a new feature at some point
from whisperx.
arnavmehta7
commented on May 22, 2024
I am using a older commit, and its now using 28gigs of GPU ram, which is TOOOO much. Any idea how to fix this?
from whisperx.
arnavmehta7
commented on May 22, 2024
Files are being processed in a loop, and seems like after every file, GPU ram usage pops up by ~2gbs ram.
from whisperx.
arnavmehta7
commented on May 22, 2024
@m-bain I just ran a very long audio and it showed that it tried to allocate 30gbs of GPU ram, is there any clean fast fix?
Thankyou!
from whisperx.
Related Issues (20)
- Should pyannote.audio be added in requirements.txt? HOT 1
- whisperx.load_model HOT 1
- pure phonetic japanese transcription as hiragana/furigana or romaji (without kanjis)
- transcribe audio with "guidance" text as reference HOT 1
- AttributeError: 'Wav2Vec2Processor' object has no attribute 'sampling_rate'
- Diarization Pipeline config on diarize.py HOT 6
- May I ask how C # can use this project
- Audio captions (background sounds, e.g. [applause], [dog barking])
- Keeping the same speaker in different files
- Question about current BSD-4-Clause License HOT 2
- Is there any way to disable VDA?
- ValueError: Requested float16 compute type, but the target device or backend do not support efficient float16 computation HOT 3
- detect_language() doesn't work on a GPU? HOT 2
- Transcription/translation language change -- '50358' / '50359' is not a valid task (accepted tasks: transcribe, translate)
- Wrong cuda device (device_id > 0) on VAD in whisperx.load_model() method
- A tip: color coding the outputs (e.g. srt) from the JSON files HOT 1
- RuntimeError: Library libcublas.so.12 is not found or cannot be loaded HOT 3
- Word Level Transcripts Error HOT 2
- Transcription fails with diarization enabled
- Some transcriptions missing properties HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from whisperx.