Modeling Real-Time Interactive Conversations as Timed Diarized Transcripts

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tanzer, Garrett, Ahdritz, Gustaf, Melas-Kyriazi, Luke
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909208184619008
author Tanzer, Garrett
Ahdritz, Gustaf
Melas-Kyriazi, Luke
author_facet Tanzer, Garrett
Ahdritz, Gustaf
Melas-Kyriazi, Luke
contents Chatbots built upon language models have exploded in popularity, but they have largely been limited to synchronous, turn-by-turn dialogues. In this paper we present a simple yet general method to simulate real-time interactive conversations using pretrained text-only language models, by modeling timed diarized transcripts and decoding them with causal rejection sampling. We demonstrate the promise of this method with two case studies: instant messenger dialogues and spoken conversations, which require generation at about 30 tok/s and 20 tok/s respectively to maintain real-time interactivity. These capabilities can be added into language models using relatively little data and run on commodity hardware.
format Preprint
id arxiv_https___arxiv_org_abs_2405_13203
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Modeling Real-Time Interactive Conversations as Timed Diarized Transcripts
Tanzer, Garrett
Ahdritz, Gustaf
Melas-Kyriazi, Luke
Machine Learning
Computation and Language
Chatbots built upon language models have exploded in popularity, but they have largely been limited to synchronous, turn-by-turn dialogues. In this paper we present a simple yet general method to simulate real-time interactive conversations using pretrained text-only language models, by modeling timed diarized transcripts and decoding them with causal rejection sampling. We demonstrate the promise of this method with two case studies: instant messenger dialogues and spoken conversations, which require generation at about 30 tok/s and 20 tok/s respectively to maintain real-time interactivity. These capabilities can be added into language models using relatively little data and run on commodity hardware.
title Modeling Real-Time Interactive Conversations as Timed Diarized Transcripts
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2405.13203