-
### The Feature
Popular ASR services that should be included - Deepgram, AssemblyAI, Google ASR, Self hosted whisper should be supported in litellm.
### Motivation, pitch
These are popular ASR sys…
-
Do you have any plan to support,
- Nvidia Riva [STT](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/asr/asr-overview.html)
- Nvidia Riva [TTS](https://docs.nvidia.com/deeplearning/riva/us…
-
### System Info
Encoder: WavLM-large
Projector:Linear(~18.88M)
LLM: vicuna-7b-v1.5,
### Information
- [x] The official example scripts
- [ ] My own modified scripts
### 🐛 Describe the bug
进…
-
Similar to https://github.com/lfortran/lfortran/issues/4686, but for ASR text output.
So that the following:
```
lfortran a.f90
```
Can also be done using:
```
lfortran a.f90 --show-asr > a.…
-
### Self Checks
- [X] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [X] I have s…
-
执行这个代码后
python syncnet.py --save_dir ./syncnet_ckpt/ --dataset_dir ./data_utils --asr huber
报错：./data_utils/aud_hu.npy
各位大佬是什么原因
-
运行python train.py --dataset_dir ./data_utils/ --save_dir ./checkpoint/ --asr hubert --use_syncnet --syncnet_checkpoint syncnet_ckpt
报错：PermissionError: [Errno 13] Permission denied: 'syncnet_ckpt'
文…
-
# Task Name
[Task name]: Target Speaker ASR
[Description]: Given a multispeaker speech utterance, decode the text corresponding to the specified speaker.
## Task Objective
Multispeaker ASR i…
-
And expose it as a tool for users:
https://fortran-lang.discourse.group/t/small-fortran-tools/8793/12
It would take a function `f(a, b, c)` and partially specialize an argument, say "a=1", inlin…
-
推理时，使用文档中的语句：
```bash
python inference.py --asr hubert --dataset ./your_data_dir/ --audio_feat your_test_audio_hu.npy --save_path xxx.mp4 --checkpoint your_trained_ckpt.pth
```
![image](https://…