automatic speech recognition (ASR)

A real-time Speech to Text (STT) engine

C# library that provides an easy to use abstraction of the Vosk speech recognition toolkit

This repository aims at archiving the code used during the performance ECPC 2022 that took place at the Paris Fine Art School with the collaboration of IRCAM.

LEM speech recognition device, designed for Signal Processing lecture.

This repository provides resources for a Quick Start guide for connecting Amazon Connect with Xdroid platform to provide post-call analytics. Intended target audience are system administrators who manage and configure the AWS Amazon Connect instance, and also for system architects and support engineers.

Models trained with Kaldi

TIMIT: famous corpus of American English with phone-level transcriptions (LDC93S1).

LibriSpeech: large ASR data set of read books (SLR12) [Panayotov et al. 2015].

AN4: Alphanumeric or "census" database from CMU [Acero 1993].

Buckeye: Buckeye Speech Corpus (release 2) of interviews from Ohio State.

Proof-of-concept (POC) app towards Aida English app.

IndicTTS: Indian English speech from the IIT TTS Team.

EMIME Bilingual {Finnish,German,Mandarin}/English database (www.emime.org).

UCAM Bilingual database from EMIME (www.emime.org).

CRM (coordinate response number) corpus [Bolia et al. 2000].

GRID: audiovisual corpus of grid-related commands, from Univ. Sheffield.

Lombard Grid: extension of Grid corpus with Lombard and normal speech.

CTIMIT: TIMIT played through cellphone network (LDC96S30).