-
Speech recognition is a standard generation task where the input is speech, output is text. For now, analysis could be done on the output side only.
* Evaluation metric: word error rate, character …
-
This project will utilize a [dataset composed of Donald Trump’s campaign speeches](https://www.kaggle.com/datasets/christianlillelund/donald-trumps-rallies) across 35 different rallies spanning from D…
-
Audio data is much smaller than video and probably also than photos, so there's a chance this fits CPU processing. Being ~linear~[edit: sequential], this may be a good fit for RNN [edit: recurrent neu…
-
### Preliminary Remark
The observations presented here are also relevant for the _polmineR repository._
### Some Background
The _Bundestag Protokolle_ often employ spacing to enhance readability …
-
# Task Name
Predicting the sentiment of given spoken utterance
## Task Objective
Sentiment analysis refers to classifying a given speech segment as having negative, neutral, or positive senti…
-
**Is your feature request related to a problem? Please describe.**
For the word models of Finnish parliamentary data, the preprocessing included lemmatization with spaCy's `fi_core_news_lg` model. Re…
-
I was doing [Dataset Cartography](https://arxiv.org/abs/2009.10795) analysis on the training dataset for the e2e SLU model based on a whisper encoder. This analysis splits the dataset into 3 parts: ea…
-
@nschneid wrote the other day regarding the analysis of reported speech in response to: @MagaliDuran, that "the policy was recently changed but not fully updated in the guidelines".
This took me b…
-
I would like to add various AWS Scripts such as S3 bucket creator, downloading from S3 bucket, creating an EC2 instance, using AWS Transcribe API for speech to text analysis, AWS Polly API for text to…
-
**Abstract**
Computers can tell us whether we’re happy, sad, angry or any of the several emotions we feel. Computers can understand what we’re saying and answer back. How does all this magic happen? …