Achieving Human Parity in Conversational Speech Recognition · HackerPapers discussions

Achieving Human Parity in Conversational Speech Recognition — mentioned by albertzeyer

Thu, 22 Oct 2020 12:15:45 GMT

This (https://arxiv.org/abs/2010.03449) is by far not the first publication which claims super human speech recognition, or human parity. For example https://arxiv.org/abs/1610.05256 or https://arxiv.org/abs/1703.02136. And since then, the WER has dropped much more on the same dataset (Switchboard), for example https://arxiv.org/abs/2001.07263. They highlight here the aspect of low latency, so this work might be current SOTA under this condition. Anyway, yes, we have reached human parity, or even surpassed it, for this dataset, under the same conditions. The conditions are: * English…

Paper and discussion history. A mention is not an endorsement.

Achieving Human Parity in Conversational Speech Recognition — mentioned by dang

Mon, 21 Aug 2017 09:12:27 GMT

Ok, we'll change to that. Thanks! From https://arxiv.org/abs/1610.05256 via https://blogs.microsoft.com/ai/2016/10/18/historic-achieveme....

Paper and discussion history. A mention is not an endorsement.

Achieving Human Parity in Conversational Speech Recognition — mentioned by lucidrains

Fri, 30 Dec 2016 17:54:08 GMT

https://arxiv.org/abs/1610.05256

Paper and discussion history. A mention is not an endorsement.

Achieving Human Parity in Conversational Speech Recognition — mentioned by szupie

Tue, 18 Oct 2016 20:09:04 GMT

Section 9 in the paper[1] is all about comparing these mistakes between the system and humans. The most common mistakes for humans and the system are in tables 9—11. We find that the artificial errors are substantially the same as human ones with one large exception confusions between backchannel words [acknowledgment words like “uh-huh”] and hesitations. The difference they found, but suspect might be a result of the different transcription guidelines of the training corpus: we see that by far the most common error in the ASR system is the confusion of a hesitation in the reference for a…

Paper and discussion history. A mention is not an endorsement.

Achieving Human Parity in Conversational Speech Recognition — HN submission

Tue, 18 Oct 2016 18:19:52 GMT

Achieving Human Parity in Conversational Speech Recognition

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Achieving Human Parity in Conversational Speech Recognition — mentioned by Eridrus

Tue, 18 Oct 2016 17:44:57 GMT

The actual paper has a section on error analysis that is particularly enlightening: https://arxiv.org/abs/1610.05256 On the CallHome dataset humans confuse words 4.1% of the time, but delete 6.5% of words, most commonly deleting the word "I". Their ASR system confuses 6.5% of words on this dataset, but only deletes 3.3% of words, so depending on how you view this their claim about being better than humans isn't definitely true, if you consider the task to be speech recognition, rather than transcription. Also, while the overall "word error rate" is lower than humans it's not clear if this is…

Paper and discussion history. A mention is not an endorsement.