24:37

Ilya Sutskever: Sequence to Sequence Learning with Neural Networks at NeurIPS 2024

31:30

2024 ZGC Forum: Dan Povey - Effect of AI on Society

11:46

#68 Questions From Youtube Viewers

08:35

#67 K2fsa: Training Advice for Beginners

13:49

#66. K2fsa: Beginner's Questions: Tensorboard, Troubleshooting

06:53

#65 Nadira: Icefall: LibriSpeech Training for Beginners

05:48

#64 Nadira: Train RNNLM and 2-gram for LODR decoding

03:37

#63 GPUs for  Model Training

06:51

#62 Context Biasing

24:27

#61 Data Preparation with prepare.sh

19:36

#60 Audio Segmentation using Text Search project

07:33

#59 Predicting Multi-Codebook Vector Quantization Indexes for Knowledge Distillation

16:06

#58 D.Povey - Cambridge Life / Candlelight dinners in wizard's type robe

13:32

#57 LibriLight/LibriHeavy   Audio Alignment and Segmentation

11:55

#56 GPU-accelerated Guided Source Separation for Meeting Transcription

03:33

#55 Zipformer and Paraformer Explained

06:01

#54 LF-MMI / ASGD Training

03:15

#53 Rust /C++/Deep Learning

06:51

#52 PyTorch vs ONNX vs NCNN

01:56

#51 Nadira: Create API with AWS API Gateway

04:48

#50 ASR Future/Advice to Master's Students/JAX/TPUs

07:07

#49 FastEmit [Google Paper]

07:22

#48 Shallow Fusion

09:04

#47 Ragged Tensors

05:40

#46 Dan Kaldi: nnet2 feature extraction diagram

05:20

#45 Dan Kaldi: nnet1 nnet2 and nnet3 explained.

07:25

#44 How does New Bing with ChatGPT work?

10:23

#43 D. Povey - simple questions about researcher's life.

07:07

#42 Teaching speech recognizers new words — without retraining [Amazon AWS AI SLT 2022 paper]

00:30

Next-Gen Kaldi Android App. Part 4/4 Chinese (Background music)

00:31

Next-Gen Kaldi Android App. Part 2/4 Chinese + English

00:24

Next-Gen Kaldi Android App. Part 1/4 Pure Chinese (No background noise）

00:55

Next-Gen Kaldi Android App. Part 3/4 Chinese with noise

03:15

Nadira#41 BART: Abstractive Summarization

06:55

Dan #40: LM Rescoring for RNN-T setup

04:03

Nadira #39: k2 icefall colab notebooks

01:59

Nadira #38 Next-gen Kaldi: decode files using sherpa server

04:56

Nadira #37 Next-gen Kaldi: sherpa server

18:37

Dan K2 #36: What is BPE and lang_bpe_500?

05:51

Nadira K2 #35 Decoding test_clean and test_other from LibriSpeech dataset.

07:27

Dan Kaldi #34 Endpointing

02:33

Dan Kaldi#33 Forced Alignment

03:24

Dan K2 #32 Multiple Datasets in Training Next-gen Kaldi

05:53

Nadira K2 #31 Hugging Face and export.py

01:01:43

Dan K2 #30 Daniel Povey BAAI 2022 Conference Full Version

00:58

Dan K2 #29 Favorite Toolkit for Students BAAI Conference P9 Q6

01:25

Dan K2 #28 RNNT and Conformer BAAI Conference P8 Q5

01:36

Dan K2 #27 Data Augmentation BAAI Conference P7 Q4

01:11

Dan K2 #26 WFST to Integrate a Language Model BAAI Conference P6 Q3

01:39

Dan K2 #25 Next-gen Kaldi vs WeNet BAAI Conference P5 Q2

01:14

Dan K2 #24 Next-gen Kaldi for Smart Phone Devices BAAI 2022 Conference P4 Q1

36:32

Dan K2 #23 Reworked Conformer Model: BAAI Conference P3

05:37

Dan K2 #22 Pruned training for RNN-T BAAI Conference P2

09:31

Dan K2 #21 Next-gen Kaldi: what is it? BAAI Conference P1

03:27

Nadira K2 #20 How to Train Icefall Model

04:19

Nadira K2 #19 Hugging Face Test LibriSpeech and GigaSpeech

01:54

Nadira K2 #18 Icefall prepare.sh

07:21

Nadira K2 #17 Install Icefall

01:06

Nadira K2 #16 Install Lhotse

01:29

Nadira K2 #15 Install Graphviz