Baki Web Dev Channel

for context, i recorded 200 samples of my voice. unfortunately, to train the tts model, i need to meet the minimum audio frequency. since i recorded these samples at midnight, the frequency of my audio is too low. because of that, i couldn’t train the tts model it’s sad, but i learned this the hard way. any advice or suggestions would be greatly appreciated

I think I know the problem but I don’t know how to fix it yet

this is where i stop, any help would be appreciated

hey guys, yesterday, i was working on my tts model, unfortunately i couldn't train my own voice/language. that's because i face some challenges, challenges like preparing data and my voice for tts model and normalizing the transcription text were tough. so, i decided to use my web development skills to automate some of my tasks. i built this web to help me translate some english datasets into amharic, and i use this translated text to read out loud (record myself) and in addition it gives me a unique id like baki001-0006 , so i can use this id to save my voice recording video. when i click "save", it generates a new id and prepares the translation at the same time. in addition, when i click the download csv button, it will create and generate clear, organized csv files. it's late now, and i might work a bit afterward. for now, have a good night! 👋

free translator ከፈለጋችሁ፡

curl --location 'https://www.ethiocloud.com/ai/Services/AjaxService.svc/TranslateText' \
--header 'Content-Type: application/json' \
--data '{"inputText":"Printing, in the only sense with which we are at present concerned, differs from most if not from all the arts and crafts represented in the Exhibition","fromLanguage":"en","toLanguage":"am"}'

Hey guys , እንዴት ናችሁ ?

ዛሬ አሪፍ ጊዜ ነው ያሳለፍኩት ፤ እንዳሰብኩት በጊዜ ባለጨርስም ብዙ ነገር ነው የተማርኩበት። ነገ ስለምቀጥለው ለዛሬ ደና እደሩልኝ

ከታዘብኩት መካከል amharic online public dataset is full of እሮሮ 😂

result 0.4 loss