Datasets:

Thomcles
/

YodaLingua-English

__key__ stringlengths 23 23	text stringlengths 22 718	language stringclasses 1 value	speaker_id stringlengths 26 26	dnsmos float64 2.4 3.74
EN_b18bd2be12f3_W000000	A lot of people really get stuck on these types of puzzles, but keep watching for one simple trick to make it much simpler. Hello Huskalinks, and welcome to day 14.	en	EN_b18bd2be12f3_SPEAKER_00	3.517
EN_b18bd2be12f3_W000001	We first split the input into two sections using interactG. Then we can parse each of the lines in the second section using a parser that parses the characters and the ASCII arrow like this.	en	EN_b18bd2be12f3_SPEAKER_00	3.484
EN_b18bd2be12f3_W000002	The first section is only one line, so we can use pattern matching to grab it. For the second section, we can use Parslist.	en	EN_b18bd2be12f3_SPEAKER_00	3.463
EN_b18bd2be12f3_W000003	We need to step through 10 iterations, and for each, we will lookup the inserted character for each adjacent pair. So let's break it down into a getPairs function, and a function g that will do the lookup.	en	EN_b18bd2be12f3_SPEAKER_00	3.538
EN_b18bd2be12f3_W000004	So G will take the pair and return a list with the new item and then also the second of the original pair. Lookup returns a maybe, so let's just do a from maybe space.	en	EN_b18bd2be12f3_SPEAKER_00	3.378
EN_b18bd2be12f3_W000005	GetPairs will then just return all adjacent items as a list of pairs. And yes, I really should be using ZipWidth for this. Anyway, the last thing we need to do is to get the popularity counts, and we can use our histo function that we developed on day 6 for this.	en	EN_b18bd2be12f3_SPEAKER_00	3.472
EN_b18bd2be12f3_W000006	And then once we subtract the highest pop count from the lowest, we're done.	en	EN_b18bd2be12f3_SPEAKER_00	3.275
EN_b18bd2be12f3_W000007	Part 2 is exactly what I feared, more iterations. This is a problem because the resulting string will grow exponentially and completely blow out the memory requirements. So we need to find another approach that doesn't use so much memory.	en	EN_b18bd2be12f3_SPEAKER_00	3.445
EN_b18bd2be12f3_W000008	The trick is to avoid doing all the insertions together, so for each pair we can keep track of which iteration it's on and just work on these in a depth first way.	en	EN_b18bd2be12f3_SPEAKER_00	3.426
EN_b18bd2be12f3_W000009	Let's start by removing the stuff we don't need from part 1.	en	EN_b18bd2be12f3_SPEAKER_00	3.057
EN_b18bd2be12f3_W000010	Then we can write a function D that takes one of these three tuples and returns the histogram for the end result.	en	EN_b18bd2be12f3_SPEAKER_00	3.285
EN_b18bd2be12f3_W000011	If we're on the last iteration, just return the second item and its population of 1. But if there are more iterations to go, use recursion to combine the results after doing the insertion.	en	EN_b18bd2be12f3_SPEAKER_00	3.483
EN_b18bd2be12f3_W000012	The way we get the inserted character is the same as before.	en	EN_b18bd2be12f3_SPEAKER_00	3.292
EN_b18bd2be12f3_W000013	To reduce appended histograms, we need to group by the character and then sum all the population counts for each one.	en	EN_b18bd2be12f3_SPEAKER_00	3.423
EN_b18bd2be12f3_W000014	Let's not forget about the very first element, which we can add into the concatenation of running D over the rest of the pairs, then combine it, and then sort by the pop count.	en	EN_b18bd2be12f3_SPEAKER_00	3.513
EN_b18bd2be12f3_W000015	If we rename that to h, then we can just uncomment the old definition of f. But before we do that, I'd like to try to make this a bit more efficient by memoising the d function.	en	EN_b18bd2be12f3_SPEAKER_00	3.444
EN_b18bd2be12f3_W000016	To do that, it's as simple as importing MemoTria,	en	EN_b18bd2be12f3_SPEAKER_00	3.307
EN_d7b6f63b3547_W000005	this segment of Lakeland News, please consider making a tax-deductible contribution to Lakeland Public Television.	en	EN_d7b6f63b3547_SPEAKER_01	3.409
EN_2249769a1550_W000000	Brainerd Police Department is searching for a man who was last seen on or around July 4th in downtown Brainerd.	en	EN_2249769a1550_SPEAKER_00	2.781
EN_2249769a1550_W000001	John Allen Greenwalt was active on his Facebook page, but family and friends haven't heard from him or seen him.	en	EN_2249769a1550_SPEAKER_00	3.471
EN_2249769a1550_W000002	The 29-year-old is described as a Caucasian, 5 foot 10 inches tall, 180 pounds, with blue eyes and brown hair.	en	EN_2249769a1550_SPEAKER_00	3.523
EN_2249769a1550_W000003	Greenwald has several tattoos on his arms and neck. He has the letters John across his left fingers and Diane on his right forearm.	en	EN_2249769a1550_SPEAKER_00	3.39
EN_2249769a1550_W000004	His family became concerned because this is unusual behavior for Greenwald to not have contact with them for several days, so they reported him missing on July 14th. Investigators are on the case. Brainerd Police has been active in the search for Greenwald. If you have any information on Greenwald, you are asked to ca...	en	EN_2249769a1550_SPEAKER_00	3.504
EN_8a77515b4b60_W000000	Hey, did you know that the view rate stayed at 25 views per hour on our latest promotion?	en	EN_8a77515b4b60_SPEAKER_00	3.343
EN_8a77515b4b60_W000001	since 10pm except for one little dip. This is happening because a lot of people were going into the video, maybe new users, which may click on the link. Usually I'd get about 400 or so views because, well, the ratio is 40%.	en	EN_8a77515b4b60_SPEAKER_00	3.487
EN_8a77515b4b60_W000002	this is normal it's okay for that and for the previous video it's looking okay just getting about five views per hour as usual and there's nothing wrong with it the two days of fake supporters already finished for that video	en	EN_8a77515b4b60_SPEAKER_00	3.292
EN_3a00b329e905_W000000	Hello everybody! This video will show you how to create an Oracle VirtualBox virtual machine and how to recover data from a virtual disk of such machine after a failure or when the virtual machine can't start.	en	EN_3a00b329e905_SPEAKER_00	3.441
EN_3a00b329e905_W000001	In order to create a virtual machine, install and run Oracle VirtualBox.	en	EN_3a00b329e905_SPEAKER_00	3.458
EN_3a00b329e905_W000002	Click the new button in the toolbar at the top of the window.	en	EN_3a00b329e905_SPEAKER_00	3.498
EN_3a00b329e905_W000003	In the wizard window enter the name of the new virtual machine.	en	EN_3a00b329e905_SPEAKER_00	3.251
EN_3a00b329e905_W000004	The operating system type that you are going to install, in our case, that is Microsoft Windows, version Windows 10 64-bit.	en	EN_3a00b329e905_SPEAKER_00	3.432
EN_3a00b329e905_W000005	Please note that you can install onto your virtual machine not only Windows, but also other operating systems, including Linux.	en	EN_3a00b329e905_SPEAKER_00	3.44
EN_3a00b329e905_W000006	Assign the memory size you'd like to give this virtual machine. It is recommended to have at least 2 GB.	en	EN_3a00b329e905_SPEAKER_00	3.531
EN_3a00b329e905_W000007	I allocate 4GB because my computer has 8GB of memory installed. To make sure the virtual machine is not going to be too hard for your PC to handle, I recommend setting its memory size at no more than 50% of the actual memory available.	en	EN_3a00b329e905_SPEAKER_00	3.425
EN_3a00b329e905_W000008	Select settings for the hard disk of the virtual machine. You can create a new one or connect an existing one. I select Create a new virtual hard disk.	en	EN_3a00b329e905_SPEAKER_00	3.367
EN_3a00b329e905_W000009	In the next window, select the virtual hard disk type. By the way, there is also an Expert mode button, clicking on which you can access more hard disk settings. Now let's get back to virtual disk types.	en	EN_3a00b329e905_SPEAKER_00	3.373
EN_3a00b329e905_W000010	A user can choose the disk format for the machine being created. By default, it is VDI. If you don't need to use the disk with other software virtualization products, you can leave the setting as it is.	en	EN_3a00b329e905_SPEAKER_00	3.413
EN_3a00b329e905_W000011	In this window, you can also set the size of the disk you are creating. The size can be anything you like, but it's better to have at least 30GB and make sure that your hard disk still has enough space after you create the virtual machine.	en	EN_3a00b329e905_SPEAKER_00	3.575
EN_3a00b329e905_W000012	That is because your virtual machine may become larger in the course of work.	en	EN_3a00b329e905_SPEAKER_00	3.257
EN_3a00b329e905_W000013	You can also choose the format, dynamically allocated or fixed size. Dynamically allocated option is more efficient in terms of saving hard disk space, as such virtual disk will grow when more data is saved there.	en	EN_3a00b329e905_SPEAKER_00	3.328
EN_3a00b329e905_W000014	The virtual machine is created and we can see it in the list on the left. All of its settings are shown on the right.	en	EN_3a00b329e905_SPEAKER_00	3.425
EN_3a00b329e905_W000015	Run the virtual machine by clicking on Start, shown as a green arrow in the toolbar. As our virtual machine has no operating system installed, we have to show the path to the installation disk. By default, it is the CD-DVD drive, in our case disk D.	en	EN_3a00b329e905_SPEAKER_00	3.489
EN_3a00b329e905_W000016	However, by clicking on this icon , you can change the path and select an ISO image or a USB drive instead.	en	EN_3a00b329e905_SPEAKER_00	3.429
EN_3a00b329e905_W000017	Watch another video on our channel to see how to create a bootable USB drive to install Windows. Find the link in the description.	en	EN_3a00b329e905_SPEAKER_00	3.505
EN_3a00b329e905_W000018	I specify the path to the ISO image I have prepared beforehand, next and after that Windows installation starts.	en	EN_3a00b329e905_SPEAKER_00	3.541
EN_3a00b329e905_W000019	Installing Windows on an Oracle VirtualBox virtual machine is similar to making a clean installation on a desktop or laptop computer, that's why I won't go into details. Just watch another video by our channel to see how to make a clean installation of Windows. You will find the link in the description.	en	EN_3a00b329e905_SPEAKER_00	3.436
EN_3a00b329e905_W000020	After the operating system is installed, you will have another that is virtual operating system that works inside the operating system of your computer. All files of this operating system, both system files and user files,	en	EN_3a00b329e905_SPEAKER_00	3.396
EN_3a00b329e905_W000021	stored on the computer's hard disk. By default, this folder is named VirtualBox.	en	EN_3a00b329e905_SPEAKER_00	3.433
EN_3a00b329e905_W000022	You can change location of the Virtual Machines folder in the program settings if necessary. To do it, go to File,	en	EN_3a00b329e905_SPEAKER_00	3.328
EN_3a00b329e905_W000023	General and specify the folder you want to place virtual machines in.	en	EN_3a00b329e905_SPEAKER_00	3.532
EN_3a00b329e905_W000024	The main files of an Oracle VM VirtualBox virtual machine are as follows.	en	EN_3a00b329e905_SPEAKER_00	3.42
EN_3a00b329e905_W000025	vbox – a settings file describing the virtual machine and its settings and actually the virtual machine disk image file vdi or vmdk, vhd, vhdx, hdd, qed, qcow, depending on the format of the virtual disk that you have selected during the installation.	en	EN_3a00b329e905_SPEAKER_00	3.483
EN_3a00b329e905_W000026	While the virtual machine is running, the program can create additional files or folders if necessary. Logs – the folder containing virtual machine log files and Snapshots – the folder containing virtual machine snapshot files, i.e. images of the virtual machine condition.	en	EN_3a00b329e905_SPEAKER_00	3.444
EN_3a00b329e905_W000027	In this video, I want to show you one trick just in case if for some reason your virtual machine is no longer working and there are important files on these disks. I will show you how to recover them.	en	EN_3a00b329e905_SPEAKER_00	3.354
EN_3a00b329e905_W000028	Though Oracle VirtualBox is a virtual machine, real data can be saved there.	en	EN_3a00b329e905_SPEAKER_00	3.396
EN_3a00b329e905_W000029	As we have already said, all files stored in virtual machine disks are located in VDI files of the virtual disk . Hetman Partition Recovery, the program for hard disk data recovery, has the function of mounting virtual disks and recovering data from such disks. You can download Hetman Partition Recovery following the ...	en	EN_3a00b329e905_SPEAKER_00	3.602
EN_3a00b329e905_W000030	In order to get access to virtual machine files, run Hetman Partition Recovery and mount the virtual machine disk.	en	EN_3a00b329e905_SPEAKER_00	3.482
EN_3a00b329e905_W000031	If there are several of them, you can mount them all at once or one by one. To mount a virtual disk with the help of Hitman Partition Recovery, click on Mount Disk in the Quick Access menu of the program. As a result, the window of choosing a virtual disk will open.	en	EN_3a00b329e905_SPEAKER_00	3.48
EN_3a00b329e905_W000032	Go to the folder containing the virtual machine and choose the necessary virtual disk file. And open.	en	EN_3a00b329e905_SPEAKER_00	3.424
EN_3a00b329e905_W000033	As you do that, the section mounted disks containing the list of mounted virtual disks will appear in the window, where all disks found by the application are shown. If you mount several disks, you will see the entire disk list there.	en	EN_3a00b329e905_SPEAKER_00	3.573
EN_3a00b329e905_W000034	Scan the disk with the application by clicking on the disk in the Disk Manager. After the analysis, the application will show the directory tree of the scanned disk.	en	EN_3a00b329e905_SPEAKER_00	3.564
EN_3a00b329e905_W000035	Find and recover the necessary files to a convenient location in your main operating system. By the way, you can use this method to move files from a virtual machine disk to the main operating system.	en	EN_3a00b329e905_SPEAKER_00	3.538
EN_3a00b329e905_W000036	The methods we have described for recovering data of an Oracle VM VirtualBox virtual machine are good for virtual machine regardless of their virtual disk format.	en	EN_3a00b329e905_SPEAKER_00	3.365
EN_3a00b329e905_W000037	I also recommend watching our previous videos on VMware Workstation and Microsoft Hyper-V virtual machines. You can find the links in the description of this video.	en	EN_3a00b329e905_SPEAKER_00	3.533
EN_3a00b329e905_W000038	That is all for now. If you like this video, hit the like button below and subscribe to our channel to see more. We'll be glad to answer any questions in comments.	en	EN_3a00b329e905_SPEAKER_00	3.424
EN_ee935178bb20_W000000	Hi, I'm Kelly Francis Corrado, and I am a singer, songwriter, and multi-instrumentalist. And today we're going to learn how to tune the guitar.	en	EN_ee935178bb20_SPEAKER_00	3.502
EN_ee935178bb20_W000001	you're going to need a guitar acoustic or electric a guitar pick	en	EN_ee935178bb20_SPEAKER_00	3.345
EN_ee935178bb20_W000002	and a phone and we're going to use the tuner Guitar Tuna. It's a free app.	en	EN_ee935178bb20_SPEAKER_00	3.369
EN_ee935178bb20_W000003	Alright, let's get started. So let's learn a little bit about what everything is on the guitar and how to use it. So first off...	en	EN_ee935178bb20_SPEAKER_00	3.396
EN_ee935178bb20_W000004	right here on the guitar is called the bridge and it holds the strings in place we follow the strings	en	EN_ee935178bb20_SPEAKER_00	3.385
EN_ee935178bb20_W000005	on the neck right here this is called the nut this also holds the strings in place this is the neck	en	EN_ee935178bb20_SPEAKER_00	3.339
EN_ee935178bb20_W000006	And then these little metal things up here are called the pegs. Let's learn the strings. So the first string is the E string. The second string is the A string.	en	EN_ee935178bb20_SPEAKER_00	3.307
EN_ee935178bb20_W000007	the third one is the D, the fourth one is the G, the fifth one is the B, and this thin one at the very end is also an E.	en	EN_ee935178bb20_SPEAKER_00	3.315
EN_ee935178bb20_W000008	as you can see the top string here the E string or the every time string is a little bit thicker also has	en	EN_ee935178bb20_SPEAKER_00	3.357
EN_ee935178bb20_W000009	a low sound. So let's follow that string all the way up the neck and find the peg that goes with it.	en	EN_ee935178bb20_SPEAKER_00	3.313
EN_ee935178bb20_W000010	For me, it's this one right here. Now, the pegs work like a spool of thread. When you tighten it,	en	EN_ee935178bb20_SPEAKER_00	3.342
EN_ee935178bb20_W000011	it gets higher. When I take it away from me or push it away from me, it goes lower.	en	EN_ee935178bb20_SPEAKER_00	3.459
EN_ee935178bb20_W000012	Now this is the E string, or the every string, so we're going to try and find the E to line up here on the tuner. So let's just pluck it and see where it's at.	en	EN_ee935178bb20_SPEAKER_00	3.368
EN_ee935178bb20_W000013	Ah, now when it's over on this side, it means the string is too low.	en	EN_ee935178bb20_SPEAKER_00	3.427
EN_ee935178bb20_W000014	If the line goes over this way it's too high. What we're trying to do is get the line just in the middle.	en	EN_ee935178bb20_SPEAKER_00	3.301
EN_ee935178bb20_W000015	So I'm gonna turn this peg towards me to tighten the string and make it higher.	en	EN_ee935178bb20_SPEAKER_00	3.264
EN_ee935178bb20_W000016	Keep going and go slow. It's no fun to snap a string.	en	EN_ee935178bb20_SPEAKER_00	3.358
EN_ee935178bb20_W000017	This app is really nice too, because it'll tell you, oh, it'll tell you right there, am I too low, too high?	en	EN_ee935178bb20_SPEAKER_00	3.401
EN_ee935178bb20_W000018	the green bar goes across then we know we're in we're in tune all right yay e string okay now we're going to do the next string the second string which is the a string	en	EN_ee935178bb20_SPEAKER_00	3.57
EN_ee935178bb20_W000019	And we're going to follow that up here to the second peg. So let's just see where it's at.	en	EN_ee935178bb20_SPEAKER_00	3.408
EN_ee935178bb20_W000020	It tells it's a little low, so we're going to pull this towards us.	en	EN_ee935178bb20_SPEAKER_00	3.268
EN_ee935178bb20_W000021	Almost there, keep going. Remember, go slow. Yay, A string. Let's double check it.	en	EN_ee935178bb20_SPEAKER_00	3.309
EN_ee935178bb20_W000022	That's going to go for me. That's going to go to my third peg up here. Let's let's pluck it and see where it's at.	en	EN_ee935178bb20_SPEAKER_00	3.298
EN_ee935178bb20_W000023	Ah, so it's a little bit this way, which means it's a little high.	en	EN_ee935178bb20_SPEAKER_00	3.208
EN_ee935178bb20_W000024	So we're gonna take this tuning peg and we're gonna turn it that way. Just a little bit.	en	EN_ee935178bb20_SPEAKER_00	3.289
EN_ee935178bb20_W000025	Yay, D-string. Let me double check. Oh, almost.	en	EN_ee935178bb20_SPEAKER_00	3.52
EN_ee935178bb20_W000026	Good, okay, so now we're going to the fourth string, this one, the get string, or the G string. Now on my guitar,	en	EN_ee935178bb20_SPEAKER_00	3.431
EN_ee935178bb20_W000027	as you can see these strings are wound this way on this side the strings are wound in reverse or the opposite so we're actually going to tune the pegs the opposite way than we have been but let me show you so let's pluck the g-string	en	EN_ee935178bb20_SPEAKER_00	3.385
EN_ee935178bb20_W000028	too low so let's follow this up find the correlating peg which is right here so it says it's too low so i'm going to go higher ah so when i turn it towards me	en	EN_ee935178bb20_SPEAKER_00	3.542
EN_ee935178bb20_W000029	As you can hear, it goes lower and it tells me to. So I want to turn it away from me.	en	EN_ee935178bb20_SPEAKER_00	3.364
EN_ee935178bb20_W000030	got it okay now these next strings the b and the e are thinner and not only are they thinner they're one strings if you look at these other strings they're like one with another one woven around it so these ones can take a little bit more pulling but these you have to be very careful with	en	EN_ee935178bb20_SPEAKER_00	3.439
EN_ee935178bb20_W000031	so we have the b string and it's telling me it's too low	en	EN_ee935178bb20_SPEAKER_00	3.49
EN_ee935178bb20_W000032	ah not towards me but away from me so I'm just gonna go slow with that	en	EN_ee935178bb20_SPEAKER_00	3.392
EN_ee935178bb20_W000033	yay b string as you can see on this app too it's super helpful it'll tell you what string that you're playing and how it's going to go in tune so let's double check that b string	en	EN_ee935178bb20_SPEAKER_00	3.486
EN_ee935178bb20_W000034	Good. Now, our last string, which is the high E string, which is the every time E string.	en	EN_ee935178bb20_SPEAKER_00	3.432

End of preview. Expand in Data Studio

YodaLingua-English

YodaLingua is a high-quality speech dataset designed for training text-to-speech (TTS) systems, ASR models, and any application requiring clean, well-aligned audio–text pairs.
This release contains the English portion of the multilingual YodaLingua collection.

🧾 Dataset Overview

Property	Value
Total clips	42,182 audio–transcription pairs
Total duration	115.1 hours
Speakers	1,810 distinct speakers
Audio format	MP3 • mono • 24 kHz • 16-bit
License	Permissive — commercial use allowed

All audio clips are noise-reduced, normalized, and matched with accurate transcriptions.

Data Fields

Each entry in the dataset contains the following fields:

Field	Description
`__key__`	Unique identifier for each sample.
`audio`	Path to the audio file associated with the sample (MP3 format).
`text`	Ground-truth transcription of the audio segment.
`language`	Language code following ISO 639 standards.
`speaker_id`	Unique identifier assigned to each speaker. Multiple audio can share the same speaker ID.
`dnsmos`	DNSMOS P.835 Overall (OVRL) score estimating perceptual speech quality; higher values indicate cleaner and more intelligible audio.

🌍 Multilingual Versions

Other languages are available in the YodaLingua multilingual collection:
👉 https://huggingface.co/collections/Thomcles/yodalingua

We apply a multi-stage pipeline to ensure maximum data quality:

1. Standardization

Convert to WAV
Mono channel
Resample to 24 kHz
16-bit sample width
Normalize to –20 dBFS (with volume correction between –3 and +3 dB)

2. Noise Reduction

Advanced denoising applied to improve clarity and remove background artifacts.

3. Speaker Diarization

Segment long recordings by speaker to improve diversity and ensure speaker-consistent utterances.

4. Voice Activity Detection (VAD)

Merge consecutive VAD segments from the same speaker into clean utterances of 3–30 s.

5. Transcription

State-of-the-art ASR models produce accurate text transcripts.

6. Quality Filtering

Clips are filtered using DNSMOS P.835 OVRL; only samples with a score > 2.4 are retained.

📚 Loading the Dataset

from datasets import load_dataset

ds = load_dataset("Thomcles/YodaLingua-English")

Contact

e-mail : cyprienoucortex@gmail.com

Downloads last month: 10

Collection including Thomcles/YodaLingua-English

YodaLingua

Collection

A collection of high-quality speech datasets with transcripts in multiple languages. • 32 items • Updated 3 days ago • 6