@sabbamitta, @aminah. As I look at this, I see that indeed this is AI in pain. Those segments are HUGE and AWS is definitely complaining. The code has two ways of shortening the text presented to AWS to observe its text limitation:
it reduces the size of the XML without loss of information.
it eliminates the XML which forced AWS to treat every word separately to avoid HIndi phrasing. The loss of intelligibility therefore arises out of Aditi’s attempts to apply Hindi phrasing to Pali.
The second measure is rarely necessary, since it is draconian and desperate, and reduces intelligibility. And that is the sad result we hear today in MN12 segment 31. I have a hope that AWS may raise its limit in the future, which would allow us to avoid using measure #2. For this reason I have moved this bug out of v0.9.1 and into the SCV Backlog. We thereby acknowledge the bug but have no practical solution at this time. Long segments will be less intelligible until this is fixed. If this issue continues to arise and be disruptive, we may have to research a costly solution in advance of any action taken by AWS. That would be a PM decision.
HRH @Aminah, long may she reign as Product Manager.
The Product Manager is the voice of the customer. The PM prioritizes user requirements and requests. In this way, dispute ceases amongst engineers except on implementation issues. This role is the heart of SCRUM. The normal term is “Product Owner”, but given Aminah’s current modest shyness ( ), she might feel uncomfortable with that. I really do think of her as the Product Owner.
And your own role I would characterize as Quality Assurance Lead of the development team. You have identified more issues than any of us. Indeed, we have trouble keeping up with you!
I’m just in charge of the “how”. It’s best that I do just this because I have been known to tilt at windmills.
Good QA Leads have a natural instinct and awareness of inconsistencies. I have worked with many such people. You too have that natural ability. And you actually don’t find fault. You find inconsistent behavior. Then you ask if it is intentional. Finding inconsistencies is much much harder than finding fault. It is harder because it requires immense awareness of internal consistency and a capacity to understand and absorb the new.
Excellent! As per your PM (private message) with the PM (HRH) from my perspective you already accomplished the primary reason for even having a v0.9.1 (fixing the “no translation” bug) and everything after that is “cheery on top” and anything big after that most definitely ought to be pushed to the backlog as you’ve done here.
You know I’m a staunch republican (UK sense), right?
Likewise, it’s the first time anyone has asked me to be a bossy-boots! I’ve decided it’s alright to wear the title because all I seem to have to do is say “okay, that sounds like a good idea” and just let Karl get on with building is incredible app.
And special thanks to @sabbamitta for her discerning ear that misses nothing.
Up next we’ll be tackling Release v1.0.0. Bhante Sujato has a Pali pronunciation boot camp that we have to run Aditi through. After v1.0.0, we shall switch to maintenance/research mode for a while investigating German translation voice support. However, if you have ideas for features, please do let @Aminah know. Aminah is the SuttaCentral Voice Product Owner.
I cleared out the sound cache and you will probably experience some slowness on suttas you have previously played. They should play at the normal speed the second play. We made so many changes that all the existing sounds in the cache are probably useless, so I cleared the cache.
So that things don’t get too disorderly, Karl and I agreed that we’d close this thread after the release of v0.9.1.
Fear not! There’s a new thread here:
I’m going to put an auto close for 20hrs so as not to deprive anyone the opportunity to offer Karl hearty congratulations for his accomplishments so far (although, really I can understand if you’re all too busy listen to the suttas to even do that ).
Again, deep thanks to both Karl and all supporting contributors!