Audio Book Excerpt: Timing, Extracts B & C (Richard Abbott)

Excerpts from Timing, Book 2 in the Far from the Spaceports series by Richard Abbott

Click image for book description
Click image for more info

Today I’m pleased to present to readers what’s next up in our series featuring author Richard Abbott, whose space jaunts have so delighted meand many others. Of course, I’d previously reviewed Abbott’s debut sci-fi novel, Far from the Spaceports, followed up by another for its sequel, Timing. The audio excerpts below come from the second novel and, like our previous entry, utilize Amazon’s Polly software, which is enabled for text-to-speech in multiple accents and intonations. This compares to Alexa, a single voice.

Before moving forward, for those unfamiliar with the novels and their plots, I’ve linked the book covers to their respective Amazon blurbs. Abbott’s world-building opens a new type of sci-fi, one accessible even to those not typically enamored of the genre (such as myself), and Mitnash, along with his partner Slate, originating from artificial intelligence (AI), will capture your imagination as they seek to solve the mysteries of high-tech crime in space. Last time we listened in as a group discussed data they’d studied; today we pick up our place within Timing at a moment when Mitnash and Slate are perplexed about Callisto, then continue as the pair are joined by Parvati and Chandrika, who share their conversation regarding the loss of Selif’s ship.

Owing to the limitations on the range of voices offered just now, as Abbott points out, one voice’s accent isn’t quite where it should be. Will you spot it? Readers with previous experience in this technology, unlike myself, are likely to agree with the author, who expresses his fascination at “just how much the field has moved on since the first ‘computer voice’ some years back.” He adds that the wording is “broadly the same as the book, but changed in a few places where it sounded more flowing to move words around (funny how different spoken-aloud can sound than read-in-your-head).” (Indeed, that is so!) I’ve indicated changes in red, with words omitted from the original text in brackets and red font, and any added text (just one spot, in this instance) in red.

Note that there is more omitted text than what I’ve indicated, and can be seen in the paperback edition on pages 38-43. Toward the end of Extract A, however, while Abbott’s choice of excerpt makes perfect sense, I added in the rest of Mitnash’s statement (Maybe. But I’m not convinced. Is there any way to see if there’s any crossover of personnel?) as a point of interest to indicate some text that does not yet transfer as smoothly within text-to-voice. However, if I left out the rest (see red below) the remaining text on its own would be unclear as to who is speaking, and the missing text would be sensed. Hence my longer addition here, where I did not do it elsewhere.

Finally, related to this technology are two articles pertaining to development, limitations and what awaits in the margins of progress, for better or worse. “Computer-dictation systems have been around for years. But they were unreliable and required lengthy training to learn a specific user’s voice[,]” writes The Economist at the start of this year (click here and see note below). By June, Baidu, Inc.’s Deep Voice 2 text-to-speech technology was being reported on following a paper presented the previous month, detailing its ability to “listen to hundreds of voices to learn certain speaking styles. After less than 30 minutes of time listening to each speaker, Deep Voice 2 then can recreate the style perfectly[.]” (Click for article.) Where do we go from here?

And without further ado, simply click the arrow to listen–and enjoy!

 Extract B

After that debacle, [we] Slate and I gave up investigations for a while, and just had fun. But eventually we both decided it was time to start work properly.

“Slate, perhaps we should go through the details of the problem on Phobos?”

She cleared the wall screen, and scattered a whole array of documents across the surface.

“Where shall we start?”

With a top-level summary of the losses, compared with the ones we saw on Callisto.”

A chart opened, with two traces spilling across it. Red for Phobos, blue for Callisto. They were mostly flat, with irregular spikes showing the discrepancy pattern. Irregular, but averaging out at more or less one a week when you looked at the big picture. Callisto came out slightly more often,  Phobos slightly less. Other than that, there really wasn’t a great deal of similarity. Different days, different amounts, different principal components.

I was missing something.

“Slate, we did leave the old code on Callisto running in parallel with the new, didn’t we?”

“Absolutely. With triggers to send an alert down to us, if ever the problem surfaced again.”

“And have the triggers fired at all?”

“Not at all Mit. Not even once.”

“But why not? We never found the root cause. Why isn’t the same problem happening every few days still?”

Slate was silent for a while.

“That’s a really good question. I have no idea. Maybe, whatever situation was causing it has gone away?”

“That would be an absurd coincidence.”

“Or maybe it was an insider job, and the person is keeping a low profile? Maybe we frightened them off?” [In which case Jo’s coding style has nothing to do with the problem.]

I wasn’t convinced, but we lacked information. [Maybe. But I’m not convinced. Is there any way to see if there’s any crossover of personnel?]

[I can request the staff roster. But remember Callisto: the records are very skimpy. I’ll ask Khufu what he can find out.]

[Meanwhile, is there any chance of getting a look at the code repository on Phobos?]

[No. I asked last night while you were asleep, and they won’t open a remote link. Not for anyone, not for any reason. You’ll have to wait until we get there.]

“So, is there any more we can do for now?”

“Not really. I can show you the same data in different charts, but you’re not going to learn anything helpful by looking at them.”

So we didn’t do that. We cleared the screens, out of habit, then Slate got on with whatever she did when I was not conversing with her.

 

Extract C

I wanted human company again, so I stretched, and went in search of Parvati. She was brewing chai as I wandered in to the kitchen. Seeing me, she doubled up the amounts, found a second mug, and arranged some savoury crackers and a red and yellow striped cake on a tray.

“Did you and Slate get anywhere?”

I shook my head.

“Total blank. The figures don’t tell us any more than the basic alert message we got from Finsbury, and they won’t let us access the code yet. There’s almost nothing we can do until we get there.”

We moved back to the bridge, and enjoyed the snack together.

“Chandrika just picked up the latest from the wreck site for Selif’s ship, if you’re interested?”

I very definitely was interested. We finished the crackers, and she sliced two generous portions of the cake.

“They’ve made available the results from the data recorders. There’s nothing at all unusual until about three minutes before the crash. At that point, Selif took the vessel’s riding lights offline, and uploaded an amendment to the nav plan. [Anyway,] The upload was completed successfully, taking only the expected lag. Except that a couple of seconds later, both recording devices ceased gathering data. At the same instant. That is unheard of.”

I looked at her.

“How did that happen?”

“The maintenance log for the recorders showed that Selif had skipped two routine services. So they highlighted that in the report, and almost immediately the manufacturer put out advisory notices, basically denying all responsibility if people ignore the recommended schedule. So the official version simply lists an open verdict.”

“Is there an unofficial version?”

She grinned.

“Of course. Chandrika, why don’t you tell them?”

“To be sure. I heard this from one of the personas on Martin’s. He works part-time with a man who’s an expert on the embedded systems in boat engines.”

I nodded. It was a highly specialised area, and one that I knew next to nothing about. But it made sense that a man with those skills would have an opinion on data recorders.

“Well, he said two things. One is that a full restart cycle for those boxes is about half a second longer than the time from the point of failure, up until the impact on Tean. And the second thing is that there are only two known exploits for that model of recorder which could bring down both boxes together. One of them cannot possibly have anything to do with this case: a different ship configuration altogether. The other one happens to rely on a routing plan change.”

I sat there, absorbing the news. It made sense that these units would go into an automatic reboot mode if they went dark for some reason. Normally, that would restore them to full operation in plenty of time to carry on doing their job. But in this case, the boat had hit Tean before they had started up again. I stirred in my seat, but Slate beat me to it.

“That’s very precise timing on someone’s part. Does anybody think it is just a coincidence?”

“Oh, Slate, the official verdict is open. Nobody is suggesting anything.”

We all laughed together.

“Either it was phenomenally bad luck on their part, or…”

I paused, and Parvati continued.

“Or else someone wanted rid of them, and found a clever way to do it.”

*********

Click here for the previous entry in this series, Extract A, and stay tuned for my review of Half Sick of Shadows as well as more from Richard Abbott!

About the author…

Richard Abbott writes fiction set in two very different places. First, there is science fiction, set in a near-future solar system exploring issues of high-tech crime and human-machine relationships. The second area is historical fiction set in the Middle East at the end of the Bronze Age, around 1200 BC.

His first science fiction book, Far from the Spaceports, introduces Mitnash Thakur and his virtual partner Slate as they investigate financial crime in the asteroid belt. Its sequel, Timing, was released in the second half of 2016.

His historical fiction books explore events in the Egyptian province of Canaan, following events in the life of a priest in the small hill town of Kephrath during a time of considerable change throughout the region.

His latest publication, Half Sick of Shadows, a retelling and metamorphosis of Tennyson’s “The Lady of Shallot,” is now available for purchase, along with his other works, on Amazon and Amazon UK.

Richard lives in London, England and works professionally in IT quality assurance.

When not writing words or computer code, he enjoys spending time with family, walking, and wildlife, ideally combining all three pursuits in the English Lake District.

*********

You can follow and learn more about Richard Abbott and his books at Facebook and Google–and be sure to check out his brilliant collection of images! The author also has some amazing content at his blog, In a Milk and Honeyed Land, including Polly and Alexa, with space-related and not, his own reviews and reading list, information about his other books, and much more.

*********

Note: The Economist link loads slowly, if at all, though I’ve noticed a direct copy/paste of address to bar, as opposed to linkage, seems to do the trick: https://www.economist.com/news/leaders/21713836-casting-magic-spell-it-lets-people-control-world-through-words-alone-how-voice

*********

Advertisements