Hacker News Reader

Comment by abetusk | Hacker News Reader

I've never even heard of VOSK-API [0], the underlying offline speech to text engine that this project uses.

Does anyone have experience using it? Is it any good?

[0] https://github.com/alphacep/vosk-api

Vosk powers Dicio, a free and open source voice assistant for Android. If you have an Android device, this app is another way to try out Vosk:

- F-Droid: https://f-droid.org/packages/org.dicio.dicio_android/

- Source: https://github.com/Stypox/dicio-android

- HN: https://news.ycombinator.com/item?id=29762526

The accuracy of the English language recognition is not bad. I'm glad to see an implementation of Vosk for desktop Linux.

can second Dicio to give Vosk a try. For a local model it worked surprisingly well. But you can't yet mix languages mid sentence - difficult when searching for restaurants that have english names but are not located in a english speaking country.

Vosk-api isn't an SST engine itself, it is built using the Kaldi speech recognition toolkit (https://github.com/kaldi-asr/kaldi) and nicely implements and packages an API for Kaldi chain/LF-MMI models.

I use it to transcribe English robocalls. Vosk gets all the words right as long as I use the "Accurate generic US English" model. PocketSphinx (with the default en-us.lm.bin model in the distro package, no idea what it is) didn't get a single word right IIRC. I didn't try anything else.

Yeah, I was really impressed with the project when I encountered it last year when trying out a bunch of FLOSS Speech-To-Text options.

It was significantly better than the other FLOSS options I looked at--both in terms of getting it going initially & the quality of the speech to text results.

I tested it with a lightly modified version of this example script: https://github.com/alphacep/vosk-api/blob/master/python/exam...

What I found particularly interesting was when you have the "partial" recognition output shown in real-time you get to see how--at the end of a sentence--it may change a word earlier in the sentence in the final recognition output based on (I guess) the additional context of the full sentence.

(I just did a quick test again (with the installs from my testing last year) using an internal laptop microphone & the test script recognized a significant chunk of my speech (using a headset definitely improves things though) whereas with the same environment a test with `mic_vad_streaming` (from `DeepSpeech-examples-r0.9` with `deepspeech-0.9.0-models.pbmm`) failed to recognize any words at all.)

It's very well known among ppl who know the field. It's quite good, the lead has a nice blog too.

Results depend heavily on which speech files you use. You can even guess which it was, looking at the errors it makes.

Reply on news.ycombinator.com