-
Hello @OmerShubi 🤗
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through AK's daily papers, the paper page is available here: https://huggingface.co/pap…
-
We want to add Viterbi decoding to PyTorch. Viterbi decoding is a well-known algorithm that finds the path of maximum likelihood over a time-varying distribution. It is used in automatic speech recogn…
-
### 🚀 The feature, motivation and pitch
Now vllm cannot support diverse beam search which transformers already supports(https://huggingface.co/docs/transformers/generation_strategies#diverse-beam-sea…
-
Hi WhisperX Team,
I was wondering if you consider support https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/whisper the tensorrt-llm backend of whisper. I have done several benchmark test us…
-
需求是一些动态页面，比如图片轮播，已经用mutable_image实现了，想部署到web上，自己试着emcc编译了一下，发现image_fill没法用，换成了bitmap_transform实现类似功能在编译通过，然后在浏览器上显示，发现什么都没出来。
问下这个可以支持吗？
-
### Motivation
Many `lmdeploy` counterparts(vllm, transformers, exllamav2...) provide `logits_processors` that allow users to modify the logits before softmax. This enables many useful features like …
-
Hi,
For hugginface, it support various sample strategy: https://huggingface.co/docs/transformers/main/main_classes/text_generation
greedy decoding by calling [greedy_search()](https://huggingfac…
-
### 🚀 The feature, motivation and pitch
Basically in decoding stage, we use sampling strategy with parameters like top_p/top_k/temp/xx_penalty to control the diversity of the generated content. Howev…
-
While decoding https://test-videos.co.uk/vids/bigbuckbunny/mp4/h264/1080/Big_Buck_Bunny_1080_10s_30MB.mp4 CABAC decoding shows up as the biggest bottleneck:
https://share.firefox.dev/36VWWkH
comp…
-
## ❓ Questions and Help
Hello MMS,
I run test on cmn-script_simplified with MMS-1B-all model, got 44% WER, which is an unacceptable result.
Audio:
[NCYzUhAtZNI_0066.zip](https://github.com/…