དུས་ཐུང་མགྱོགས་ལས་དང་པོའི་གྲུབ་འབྲས་འདོན་རྒྱུ།

[TsewangJamling]

ལས་ཀ། དབུས་སྐད་ཡིག་འབེབས་རྒྱུ་ཆ་7000ལྷག་ཙམ་ཞུ་དག་ཟིན་པའི་རྗེས་རིག་ནུས་ཀྱི་ཡིག་འབེབས་སྤུས་ཚད་ལ་ཚོད་བལྟ་བྱེད་རྒྱུ།
འགན་འཁུར་བ། མཉེན་ཆས་ཆེད་ལས་པ། @gangagyatso4364
ལས་གྲོགས།
དུས་བཀག 13/01/2025 to 26-01-2025

[TsewangJamling]

After completing the transcription and its review of about 7000 audio tasks in two weeks, my team is supposed to test the ability of the AI and see how it performs.
So far I have not witnessed a single test result, of Colloquial Tibetan Speech to Formal Tibetan Text, which I've been told by our inhouse TechTeam that the results were not very encouraging the last time they did it. And also that it takes proper planning and strategy, (Benchmark Data) before running a test and also that it costs a good fortune!
Me and my team on our part, have little knowledge about the workings of the AI machine or test strategies, therefore, it would be better if the TechTeam could tell us how much more data do you need (in task numbers or audio minutes) so that you would be ready to run the test and present me the result. Only then I can further show the results to my clients and take their views and fulfil their demands. With these dilemmas, I'm not sure how to treat this particular issue. Kindly guide. Thank you.