A 2019 Guide for Automatic Speech Recognition

Re:A 2019 Guide for Automatic Speech Recognition by skykooler on Wednesday September 04, 2019 @11:55AM (#20878420)

So how can one actually use one of these systems? I'm familiar with pocketsphinx, where you download it and then run procketsphinx_continuous and it prints a transcription of the microphone. Or Julius, where you write a grammar and run it with that and it prints the output. The process for DeepSpeech seems to be along the lines of "get access to a machine with ten GPUs, find a huge dataset, train it, and then run that somehow" - are there compiled models that can actually be shipped as part of an application?

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by jamesonthecrow on Wednesday September 04, 2019 @12:30PM (#20878796)

Obviously the big cloud players offer their own APIs and SDKs (for a price), but there are a few other solutions worth looking at.

Facebook has open sourced some pre-trained models: https://github.com/facebookresearch/wav2letter

Picovoice has some smaller, more efficient models capable of running on edge devices: https://github.com/Picovoice

Full ASR does require quite large models and datasets, but you don't need nearly that much power or data to fine-tune a model for your own domain.

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by GordonS on Wednesday September 04, 2019 @01:40PM (#20879548)

Wasn't aware of Picovoice. Just tried the live do they have on their website... wow, it's... not great! Even if I spoke as precisely as possible or/and put on an American accent, it was way off the mark.

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by skykooler on Wednesday September 04, 2019 @01:27PM (#20879408)

Ah, thanks, I wasn't aware Picovoice had released an actual engine yet.

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by TaylorAlexander on Wednesday September 04, 2019 @05:26PM (#20881600)

Mozilla is maintaining an implementation of Baidu's deepspeech with pretrained models (and checkpoints so you can fine tune with your own dataset).

They have an example that accepts streaming from the microphone: https://github.com/mozilla/DeepSpeech/tree/master/examples/m...

See the last full release here: https://github.com/mozilla/DeepSpeech/releases/tag/v0.5.1

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by lunixbochs on Wednesday September 04, 2019 @02:17PM (#20879935)

I made a realtime mac demo of the acoustic model part of wav2letter++ including several trained models:

https://github.com/facebookresearch/wav2letter/issues/327

Someone in that thread ported it to linux. This demo is just "acoustic model emissions", which are character level predictions (with no repeated characters), but I have "decoding" (turning into english sentences) working locally as well and I'll post a new demo at some point.

[ Parent | on HN ]

Re:A 2019 Guide for Automatic Speech Recognition by nwalker85 on Wednesday September 04, 2019 @12:28PM (#20878773)

Verbio offers an ASR with built-in grammars, which is what you are asking for. Deepspeech is hot garbage in my experience.

[ Parent | on HN ]