Blizzard 2019 Heap

	A	B	C	D	E	F	G
1
2
3			CMU's Framework
4
5
6	Sl No	Date	Category	Task	Status	Module
7	1	23 December 2018	Development	Add forward_incremental	DONE	Vocoder
8	2	23 December 2018	Development	Add local and global conditioning	DONE	Vocoder
9	3	24 December 2018	Infrastructure	Setup on ADA	DONE	Vocoder
10	4	24 December 2018	Research	Add pre and post conv layers	stack	Vocoder
11	5	24 December 2018	Development	Generate output and compare	DONE	Vocoder
12	6	24 December 2018	Development	Check the wiring of loss function	DONE	Vocoder
13	7	24 December 2018	Development	Setup to sample instead of max	DONE	Vocoder
14	8	24 December 2018	Infrastructure	Backup the checkpoints on k14	DONE	Vocoder
15	9	24 December 2018	Development	Add numpy.choice with probability picking for sampling instead of gumbel softmax	DONE	Vocoder
16	10	25 December 2018	Research	Just give it ccoeffs and ask it to predict the quantized wavefile	stack	Vocoder
17	11	25 December 2018	Research	Code a pitch mark version of WaveNet	DONE	Vocoder
18	12	25 December 2018	Development	Add batching to pitch mark version	stack	Vocoder
19	13	26 December 2018	Debugging	Generate output per pitch mark and compare	DONE	Vocoder
20	14	26 December 2018	Debugging	Double check the eval mode	DONE	Vocoder
21	15	26 December 2018	Impact	Add an RL agent to decide which or how much data to pick	stack	Vocoder
22	16	26 December 2018	Investment	Add reparameterization based on original signal ( true posterior)	stack	Vocoder
23	17	26 December 2018	Infrastructure	Upgrade the website to a table instead of flow chart based representation for samples	stack	Documentation
24	18	26 December 2018	Development	Add tqdm to the scripts for indicating progress	stack	Generic
25	19	26 December 2018	Research	Add windowing to the file before feeding to the model	stack	Vocoder
26	20	26 December 2018	Research	Replicate pitch mark expt for SLT and generate plots, files	DONE	Vocoder
27	21	26 December 2018	Development	Setup SLT pitchmark voice on ADA	DONE	Vocoder
28	22	26 December 2018	Investment + Impact	Write about Speech Synthesis like this : https://www.youtube.com/watch?v=OWJCfOvochA	stack	Generic
29	23	30 December 2018	Debugging	Debug using Image Captioning	DONE	Vocoder
30	24	30 December 2018	Investment	Do some more expts with pad pack sequencing	stack	Generic	https://stackoverflow.com/questions/51030782/why-do-we-pack-the-sequences-in-pytorch
31	25	31 December 2018	Debugging	Debug with a single CNN layer	RUNNING	Vocoder
32	26	04 Jan 2019	Development	Finish debugging the wavenet model	RUNNING	Vocoder
33	27	15 Jan 2019	Infrastructure	Setup K13	RUNNING	Generic
34	28	15 Jan 2019	Development	Setup Acoustic Modeling on k13	stack	Acoustic Model
35	29	17 Jan 2019	Investment	Check https://www.inf.ed.ac.uk/teaching/courses/asr/2017-18/asr12-seq.pdf	stack	Vocoder
36				Quantify the goodness of fit in the context of mixture sampling	stack
37				Build Vocoder	DONE
38				Remove dc offset and build voice	stack
39				Remove silences and build voice	DONE
40				Build voice with IITM data added	RUNNING
41				Build voice with a different frame rate ( more frames)	stack
42				Fine tune by adding unrelated data	stack
43				Do Iterative Dilated Convolutions	stack
44		22 March 2019	Diagnosis	Build voice with RMS Tacotron	stack
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100