6 ms·

Going along with this: What are the latest and greatest open source speech-to-text models and/or tools out there? Would love to hear from experienced practitio

by jalopy 5y ago

Going along with this: What are the latest and greatest open source speech-to-text models and/or tools out there? Would love to hear from experienced practitioners and a bit of detail on the experience. Thanks HN community!

orra 5y ago
Mozilla announced Deep Speech[1] around the same time as Common Voice. Mozilla Deep Speech is an open source speech recognition engine, based upon Baidu's Deep Speech research paper[2]. Unsurprisingly, Deep Speech requires a corpus such as... Common Voice. [1] https://github.com/mozilla/DeepSpeech https://github.com/mozilla/DeepSpeech [2] https://arxiv.org/abs/1412.5567 https://arxiv.org/abs/1412.5567
- rasz 5y ago
  They killed this after Nvidia grant.
  - orra 5y ago
    Ah, damn. Didn't realise. It also looks like Baidu are now developing their Deep Speech as open source? https://github.com/PaddlePaddle/DeepSpeech https://github.com/PaddlePaddle/DeepSpeech
zerop 5y ago
Vosk is my favourite. I have used deep speech too. Vosk works better.
- nshm 5y ago
  Thank you. I deeply appreciate you mention our efforts. We spend quite some time and knowledge to build accurate speech recognition. Not that easy to get as much mentions as Mozilla, so we are thankful for every single one!
  - zerop 5y ago
    Vosk just works good and it works on mobile platforms too. One suggestion is to put lisence on alphacephei site. GitHub repo has it, but site doesn't.
thom 5y ago
Same question for text-to-speech!
kcorbitt 5y ago
I've had good results with https://github.com/flashlight/flashlight/blob/master/flashlight/app/asr/README.md https://github.com/flashlight/flashlight/blob/master/flashli.... Seems to work well with spoken english in a variety of accents. Biggest limitation is that the architecture they have pretrained models for doesn't really work well with clips longer than ~15 seconds, so you have to segment your input files.
mazoza 5y ago
https://github.com/coqui-ai/STT https://github.com/coqui-ai/STT
jononor 5y ago
Have used VOSK a bit recently. The out-of-the-box experience was great compared to earlier projects (looking at you Kaldi and Sphinx...). Word-level audio segmentation was one usecase, https://stackoverflow.com/a/65370463/1967571 https://stackoverflow.com/a/65370463/1967571
- woodson 5y ago
  Vosk is built on Kaldi.
- stegrot 5y ago
  Kdenlive supports automatic subtitles created with VOSK now btw. This makes it a lot more accessible for non-tech folks.
woodson 5y ago
NVidia NeMo: https://github.com/NVIDIA/NeMo https://github.com/NVIDIA/NeMo
blackcat201 5y ago
I created edgedict [0] a year ago part of my side projects. At that time this is the only open source STT with streaming capabilities. If anyone is interested the pretrained weights for english and chinese is available. [0] https://github.com/theblackcat102/edgedict https://github.com/theblackcat102/edgedict
- lazyresearcher 5y ago
  Kaldi and DeepSpeech both support streaming, right?

by jalopy 5y ago

orra 5y ago

Mozilla announced Deep Speech[1] around the same time as Common Voice. Mozilla Deep Speech is an open source speech recognition engine, based upon Baidu's Deep Speech research paper[2]. Unsurprisingly, Deep Speech requires a corpus such as... Common Voice. [1] https://github.com/mozilla/DeepSpeech https://github.com/mozilla/DeepSpeech [2] https://arxiv.org/abs/1412.5567 https://arxiv.org/abs/1412.5567

rasz 5y ago
They killed this after Nvidia grant.
- orra 5y ago
  Ah, damn. Didn't realise. It also looks like Baidu are now developing their Deep Speech as open source? https://github.com/PaddlePaddle/DeepSpeech https://github.com/PaddlePaddle/DeepSpeech

zerop 5y ago

Vosk is my favourite. I have used deep speech too. Vosk works better.

nshm 5y ago
Thank you. I deeply appreciate you mention our efforts. We spend quite some time and knowledge to build accurate speech recognition. Not that easy to get as much mentions as Mozilla, so we are thankful for every single one!
- zerop 5y ago
  Vosk just works good and it works on mobile platforms too. One suggestion is to put lisence on alphacephei site. GitHub repo has it, but site doesn't.

thom 5y ago

Same question for text-to-speech!

kcorbitt 5y ago

I've had good results with https://github.com/flashlight/flashlight/blob/master/flashlight/app/asr/README.md https://github.com/flashlight/flashlight/blob/master/flashli.... Seems to work well with spoken english in a variety of accents. Biggest limitation is that the architecture they have pretrained models for doesn't really work well with clips longer than ~15 seconds, so you have to segment your input files.

mazoza 5y ago

https://github.com/coqui-ai/STT https://github.com/coqui-ai/STT

jononor 5y ago

Have used VOSK a bit recently. The out-of-the-box experience was great compared to earlier projects (looking at you Kaldi and Sphinx...). Word-level audio segmentation was one usecase, https://stackoverflow.com/a/65370463/1967571 https://stackoverflow.com/a/65370463/1967571

woodson 5y ago
Vosk is built on Kaldi.
stegrot 5y ago
Kdenlive supports automatic subtitles created with VOSK now btw. This makes it a lot more accessible for non-tech folks.

woodson 5y ago

NVidia NeMo: https://github.com/NVIDIA/NeMo https://github.com/NVIDIA/NeMo

blackcat201 5y ago

I created edgedict [0] a year ago part of my side projects. At that time this is the only open source STT with streaming capabilities. If anyone is interested the pretrained weights for english and chinese is available. [0] https://github.com/theblackcat102/edgedict https://github.com/theblackcat102/edgedict

lazyresearcher 5y ago
Kaldi and DeepSpeech both support streaming, right?