Welcome to Reciprosody
Reciprosody is an open repository of prosodically annotated speech.
Currently, prosody researchers lack a simple way to share their annotated data. This lack:
- limits our ability to test robust supervised techniques for automatic prosody detection
- makes it difficult to compare performance across publications that operate on private corpora
- puts the burden on the creators of a resource to maintain and distribute their data.
The lack of shareable data is particularly worrisome in our field, since prosodic annotation, as we all know, is a very onerous and time-consuming task.
This material is based upon work supported by the National Science Foundation under Grant Number #1205445. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
Reciprosody is developed at the Speech Lab @ Queens College and hosted at Reciprosody.org
The code is written and maintained by Syed Reza.
The Speech Lab @ Queens College is run by Dr. Andrew Rosenberg who presides over this project. The following is an excerpt from Dr. Rosenberg's original proposal:
Prosody researchers at CUNY, Columbia and MIT are developing Reciprosody, a web source of prosodically annotated data and tools for analysis. Reciprosody makes critical resources for the study and teaching of prosody available to a growing and diverse population of researchers, students, and technologists. This website hosts data in any language, annotated under any prosody standard. The only restriction for hosting data in the repository is that it must be made freely available for academic and educational activities.
Reciprosody is not merely a passive repository, but an active resource for researchers and educators to communicate with each other on a regular basis. This enables the resource to be a center not only for annotated data, but information and support related to ongoing prosody research.