Coqui STS model data

[Linux for blind general discussion]

Hello,

when it comes to STT, there are multiple really interesting open-source
projects going on.


aside Coqui STT, there are Mozilla's deep speech, Vosk and NVIDIA's Nemo.


I've read an article recently:

https://medium.com/@nick.nagari/comparing-4-popular-open-source-speech-to-text-neural-network-models-92676a9f9265


comparing 4 popular OSS speech recognition frameworks, and Nemo with
quartzNet came out as the best choice.


 From its repository:

https://github.com/NVIDIA/NeMo


It seems really interesting.


Best regards


Rastislav


Dňa 13. 2. 2022 o 15:03 Linux for blind general discussion napísal(a):
> Apparently I need to look a bit harder  :)
>
>
> I find working  instruction here
>
> https://petewarden.com/2021/12/27/how-to-get-started-with-coquis-open-source-on-device-speech-to-text-tool/#content
>
> Anyway these STS is looking awesome..
>
> Regards
> On 2/13/22 20:10, Edhoari Setiyoso wrote:
>> Hi list,
>>
>>
>> I want to experiment with Coqui STS and need some advice.
>>
>> I download sample code from
>>
>> -- https://github.com/coqui-ai/STT-examples
>>
>> First I tried the python mic_mic_vad_streaming/ but when I ran it I've
>> lost all audio and control. I had to go to tty1 and reboot my system.
>> And then tried the nodejs_mic_vad_streaming/
>>
>> It asks for model data which I try to fetch from
>> https://coqui.ai/english/coqui/v1.0.0-digits
>>
>> I already use both my email address and still got nothing in my inbox.
>>
>> Can anyone can shed light on what to do?
>> I'd appreciate any help.
>>
>> Best Regards
>>
>>
>> Edhoari Setiyoso
>>
> --
> Edhoari Setiyoso
>
> _______________________________________________
> Blinux-list mailing list
> Blinux-list at redhat.com
> https://listman.redhat.com/mailman/listinfo/blinux-list