yeyupiaoling/Whisper-Finetune
Fine-tune the Whisper speech recognition model to support training without timestamp data, training with timestamp data, and training without speech data. Accelerate inference and support Web deployment, Windows desktop deployment, and Android deployment
CApache-2.0
Issues
- 0
Push to Hub Support
#78 opened by sanchit-gandhi - 0
- 0
whisper全量微调相关问题
#76 opened by wangyarududu - 0
tools/create_wenetspeech_data.py报错
#75 opened by Larry-Ye - 0
是否支持mp4 视频
#74 opened by lizhanyang505 - 0
Belle-whisper-large-v3-zh 没法输出标点符号
#73 opened by jianchaozhuang - 1
微调方言,数据标注为chinese还是新建一个方言的类型呢?
#72 opened by veritastry - 2
NaN or Inf found in input tensor
#70 opened by GoldenLinlin - 1
建議您整理一下 requirements.txt 及 readme.md
#71 opened by bardenthenry - 2
使用large-v2-finetune,高機率出現重複內容…
#67 opened by zero-zen - 1
能結合 pyannote 嗎?
#62 opened by HKLee2040 - 1
如何导出pt格式的模型?
#69 opened by jinchao123 - 1
训练出的模型如何导出npz格式
#68 opened by jackwenshann - 6
evaluate运行时出现问题
#56 opened by yfwang214 - 1
- 8
ubuntu20.04环境问题
#66 opened by happylsr - 2
transcribe时如何产生时间戳?
#64 opened by Koyonyong - 1
微调模型保存时需要连接huggingface
#63 opened by qiangduscu - 1
- 1
能提供fine-tune 模型的原始 checkpoint (pt) 吗?
#65 opened by walletiger - 7
- 1
请问如何导出onnx
#61 opened by drilistbox - 1
请问加速推理可以离线使用吗
#57 opened by 20246688 - 0
[CONTRIBUTION] Speech dataset Generator
#58 opened by davidmartinrius - 2
用BELLE-2/Belle-whisper-large-v2-zh识别中文音频,效果还不如Systran/faster-whisper-large-v2?
#54 opened by drilistbox - 1
可以使用initial_prompt做微调吗
#53 opened by v-yunbin - 2
使用 ct2 轉換後掉辨識率
#52 opened by ken19980727 - 2
可以单独导出我们微调模型训练出的数据吗
#50 opened by jackwenshann - 1
可以微调新的语种吗?
#48 opened by kli017 - 1
- 28
whisper large v3 Fine-Tune 後變得不太能辨識語音
#43 opened by bardenthenry - 2
- 4
如何转换V3版本
#47 opened by xyx361100238 - 1
資料格式的 language 設定
#46 opened by ken19980727 - 1
config.json文件与huggingface上的config.json不一样
#45 opened by hahazei - 5
微调时的奇怪问题,训练集变大之后,准确度反而下降了
#44 opened by ILG2021 - 5
微调在WhisperProcessor.from_pretrained调用时就报错
#42 opened by lichq5 - 15
2卡训练速度比单卡训练快很多
#40 opened by pcqiao0 - 9
正常数据和空数据一起训练的格式
#39 opened by xyx361100238 - 8
训练过程占用显存过高的问题
#30 opened by xyx361100238 - 2
如何随机化模型参数,从头开始训练
#41 opened by xyx361100238 - 1
训练稳定后的loss,大家一般都是多少啊
#38 opened by zouzoutingting - 1
一般需要多少数据,可以有不错的效果呢
#33 opened by zouzoutingting - 4
关于DataCollatorSpeechSeq2SeqWithPadding的一处问题
#32 opened by ILG2021 - 4
- 0
双通道的数据,还是单通道
#36 opened by zouzoutingting - 0
训练数据里,需要提前把标点符号去掉吗
#37 opened by zouzoutingting - 1
"无语音数据训练"是什么训练方式啊
#31 opened by zouzoutingting - 1
lora微调large V2版本,需要多大的显存,
#34 opened by zouzoutingting - 2
使用lora微调时遇到的奇怪问题
#29 opened by ILG2021