Add sentiment data preparation for IEMOCAP

[YushiUeda]

This PR adds data preparation scripts for converting emotion labels into sentiment labels ("Positive", "Negative", and "Neutral").

[codecov]

Codecov Report

Merging #4065 (f6cde1c) into master (18ca099) will increase coverage by 0.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #4065      +/-   ##
==========================================
+ Coverage   80.93%   80.94%   +0.01%     
==========================================
  Files         435      435              
  Lines       37389    37425      +36     
==========================================
+ Hits        30259    30294      +35     
- Misses       7130     7131       +1     

Flag	Coverage Δ
test_integration_espnet1	67.13% <ø> (-0.11%)	⬇️
test_integration_espnet2	52.44% <ø> (-0.06%)	⬇️
test_python	66.61% <ø> (+0.02%)	⬆️
test_utils	24.45% <ø> (-0.04%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...pnet/nets/pytorch_backend/transformer/embedding.py	97.24% <0.00%> (-0.09%)	⬇️
espnet2/bin/pack.py	95.55% <0.00%> (+0.31%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 18ca099...f6cde1c. Read the comment docs.

[ftshijt]

Looks very cool! Many thanks, will merge it