Comments (4)
I've run into this as well- one unblock I found (haven't tracked why this is the case), is that if you also include return_language=True in your pipe (so have both return_language=True, return_timestamps="word"), then the word level timestamps are correct / make sense. We were seeing some pretty nonsense timestamps without this, it could be the case that some other intermediate reps as needed to properly time align, and are only getting passed through when language info is being passed
from transformers.
Thank you for your response.
Even after I added return_language=True, the issue still persists.
This parameter does not affect the problem I've encountered.
I've run into this as well- one unblock I found (haven't tracked why this is the case), is that if you also include return_language=True in your pipe (so have both return_language=True, return_timestamps="word"), then the word level timestamps are correct / make sense. We were seeing some pretty nonsense timestamps without this, it could be the case that some other intermediate reps as needed to properly time align, and are only getting passed through when language info is being passed
from transformers.
also cc @ylacombe
from transformers.
Any update?
from transformers.
Related Issues (20)
- Trying to stack tensors from different devices in `_pad_to_max_length` in Whisper batched inference HOT 2
- [Whisper] Word-level timestamps broken for short-form audio HOT 2
- [BUG] Load StarCoder2 AWQ using Transformers HOT 6
- `import transformers` accidentally initializing both torch and jax/xla at startup time HOT 5
- FSDP Doesn't Work with model.generate() HOT 2
- Nondeterministic behavior from GPT with MPS backend HOT 6
- LlamaRMSNorm() Dtype Casting Error HOT 1
- Trainer do not move the model to GPU when doing evaluation with FSDP
- [i18n-PL] Translating docs to Polish HOT 3
- PEFT models donot "override" user's argument for return_full_text. HOT 2
- There is a probability that a bug will be triggered when tracing the llama model: torch.fx.proxy.TraceError: symbolically traced variables cannot be used as inputs to control flow HOT 1
- Couldn't connect to `https://huggingface.co`. HOT 1
- MPS memory leak?
- BLOOM embeddings should specify padding_idx HOT 5
- Importing `CLIPVisionModelWithProjection` crashes with `AttributeError: 'NoneType' object has no attribute 'dumps'` HOT 3
- 'CLIPEncoder' object has no attribute '_gradient_checkpointing_func' HOT 2
- Unexpected behavior in DonutProcessor.token2json with strings containing multilines (\n) HOT 1
- Using SparseAdam with LLaMA HOT 1
- Jamba-v01 Model + Deepspeed Zero3 lead to "RuntimeError: Detected mismatch between collectives on ranks."
- Reporting a vulnerability HOT 2
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 transformers.