LRY1994

How do the prefix parameters of prefix_tunning initialize?

1

Hello! How do the prefix parameters of prefix_tunning initialize? with random initialization ? or real words ? How could I initialize the prefix parameters on my own ?

How to implement BartAdapterModel with input_ids=None ?

1

## Environment info - `adapter-transformers` version:3.0.0 - Platform:unbuntu - Python version: 3.8.0 - PyTorch version (GPU?): ## Details # self inherits from BartAdapterModel outputs = self.model( input_ids=None, inputs_embeds=inputs_embeds ) Here...

AttributeError: 'BartConfig' object has no attribute '_to_dict_new'

2

## Environment info - `adapter-transformers` version:3.0.0 - Platform: Unbuntu - Python version: 3.8 - PyTorch version (GPU?): 1.11.0 - Using GPU in script?: yes - Using distributed or parallel set-up...

CONTROLLABLE TIME-DELAY TRANSFORMER FOR REAL-TIME PUNCTUATION PREDICTION AND DISFLUENCY DETECTION 提出联合建模标点预测和 disfluency detection，请问代码里面哪里有disfluency detection的部分？

增加LM的疑问

1

Add language model: set --lm_path in the convert_start_server.sh 这里是怎么把LM 加入decode的？

BartForQuestionAnswering cannot train adapter

1

## Environment info - `adapter-transformers` version: - Platform: - Python version: - PyTorch version (GPU?): - Tensorflow version (GPU?): - Using GPU in script?: - Using distributed or parallel set-up...

粤语识别出subword

2

## 🐛 Bug 识别出来subword 茂名口音， gt : 好 啲 呢 我 觉 得 pred: ho@@ al@@ ding ne@@ un@@ qu@@ ar@@ ter a [2-28-2_00751262_00752898.zip](https://github.com/alibaba-damo-academy/FunASR/files/14522289/2-28-2_00751262_00752898.zip) ### To Reproduce ```python model =...

assert x.size(2) == self.size AssertionError

2

https://github.com/alibaba-damo-academy/FunASR/issues/1478 https://www.modelscope.cn/models/dengcunqin/speech_paraformer-large_asr_nat-zh-cantonese-en-16k-vocab8501-online/summary model_name_or_model_dir="dengcunqin/speech_paraformer-large_asr_nat-zh-cantonese-en-16k-vocab8501-online" model_revision="master" torchrun \ --nnodes 1 \ --nproc_per_node ${gpu_num} \ funasr/bin/train.py \ ++model="${model_name_or_model_dir}" \ ++model_revision="${model_revision}" \ ++train_data_set_list="${train_data}" \ ++valid_data_set_list="${val_data}" \ ++dataset_conf.batch_size=64 \ ++dataset_conf.batch_type="token" \ ++dataset_conf.num_workers=4 \ ++train_conf.max_epoch=50 \...

请问可以提供下/data/zh_iwslt的原数据及处理过程吗？原数据链接已经失效了。 还有以下两个是来源于什么数据集？ /data/zh_pfdsj_3punc_bak /data/zh_pfdsj_3punc 关于中文标点预测数据集，目前只看到iwslt2012和人民日报2014，还有其他吗

eyError: 'asr-inference is not in the pipelines registry group auto-speech-recognition. Please make sure the correct version of ModelScope library is used.'

1

inference_16k_pipline = pipeline( task=Tasks.auto_speech_recognition, model='iic/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-offline' ) # waveform, sample_rate = soundfile.read("/data/linry/FunASR/asr_example_粤语.wav") rec_result = inference_16k_pipline(audio_in="/data/linry/FunASR/asr_example_粤语.wav") print(rec_result) 2024-02-26 18:49:21,770 - modelscope - WARNING - ('PIPELINES', 'auto-speech-recognition', 'asr-inference') not found in ast index...