Swara Identification Exercises, Vocal: Post Answers Here

[vasanthakokilam]

Thread starter for posting answers and discussions about the answers for the Swara Identification Exercises

[arunk]

Sample 1:
Ref: post #1 in the Swara Identification Exercises - Questions thread

I will kickstart with some answers.

1. Raga is Abhogi. The aro/avaro is S R2 G2 M1 D2 S | S D2 M1 G2 R2 S
2. The singer is M.S. Sheela

3. Some of the spots. Hopefully I got these right. (Note: i am avoiding the first phrase as that is part of another exercise). R2 (tara): end of 1:00 and briefly on 1:01
D2 (madya): 38s - 40s, 45-46s, 51s: basicallly D2 is used an anchor around here.
Sa (tara): 01:13s - 01:18s
R2 (madhya): 02:18, also violin between 02:19-02:20
G2( madhya): 20s (with kampita gamaka), also immediately by violin 21s-22s;
M (madhya): 16s (deep kampita with ga), again violin responds , Flat ones at around 02:02-02:03
G2 (tara): very briefly at 0:12 ta-na-na-ri . The ta-na-na-ri is d-g'-d'-/s (da, tara-ga, slide down to back to da, and slide up to tara-sa)

I will let others post their impressions before posting other answers.

[Suji Ram]

Sample 1

1. rAga- abhOgi SR2G2M1D2S/SDMGRS
2. Singer- female
3.
Mandara stayi- Between 2:41-2:53 probably touching low M1 at 2:47.

Tara stayi predominantly between 1:36-1:39
1:57-2.00, 2:08-2:10, 0:58-1.00

Madhya stayi- Starting 0:03-0:45 sec (at least) is all in middle octave. 2:25-2:40 is another example.

4. Break-down swaras:
MDS ~D, MMM mDM, R ~G M,, G RR

5.
G gamakas
0:07, 0:15, 0:19 (downward oscillation) 1:37, (high G), 2:04, 2:15-2:16

Gamaka for G is rm,g m,g (upward oscillation)
It is hard to point out downward oscillation perhaps 0:19 is an eg:

D gamaka
0.03, 0:33, 0:45, 0:50, 1:03, 1:28, 2:04-2:06

Gamaka for D: ds,ds,d It does have jaru type gamaka

[Suji Ram]

Ok, I see that the question 3 was modified later which I did not see.
So I answered as to the originally posed question -identify the the 3 octaves.
I think it is still valuable for novices like me.

[Suji Ram]

Arun,
you are spot on "tara R" 1.01 sec. It goes as RSD

[vasanthakokilam]

Let me just register this one for now: 1:13 to 1:18 - tara sa

I guessed it was M.S. Sheela as well.

On the raga, I was initially confused between Sri Ranjani and Abhogi but then settled on Abhogi.

BTW, what sruthi does she sing in? 5 or 6 kattai?

I did not recollect the Abhogi Aro/Ava and I did not want to look it up either. I do not think I got the full thing right but felt the Madhyamam to be a significant swara. The fact that Pa is not there was feelable too to some extent.

My current challenge is: There are many occasions in that short alapana where Abhogi reveals itself so beautifully in a very short interval. I want to get those including the nuances that make it the Abhogi it is.

Another thing I noticed is, the same melodic sequence in the madhya sthayi and mandhra sthayi give off different effects or 'color's. I guess I knew that before as a concept but in this attempt to get at the swaras ( though I have not made much progress yet ), the melody is so captivating that I did not even realize that some of those are the same swara sequences except in different octaves.

[arunk]

Sample 1:

4. My breakdown of swaras for the first phrase of alapana
M D S' D~ , M M M /D M (r) G~ (r) /M G R R ,

(My) Legend:
' (forward-tick i.e. apostrophe) => tara stayi indicator
` (back-tick) => mandra stayi
~ => kampitam
() => anuswara (i am not always consistent with it)
/ => slide-up from previous-swara/anuswara
=> slide-down from prev-swaraanuswara

[arunk]

BTW, what sruthi does she sing in? 5 or 6 kattai?

(Using an online keyboard) It seems to be G# and so 5.5

Arun

[arunk]

Arun,
you are spot on "tara R" 1.01 sec. It goes as RSD

Suji - I wonder if we may be looking at slightly different timespots. Between 00:57-0:58 (itunes) to about 1:01, she sings

ta - da - ri - na - na - na - aa.. - ri.. - ri - ri

(To me, this seems like )

d - s' - r - g' - r' - r' - (d)g', - r', - s' - s'

Can you confirm?

I basically indicated the later last ri there, and you may have indicated earlier ones?

(Assuming I am right), This also indicates how many swaras figure within a matter of seconds and should tell us that construction of swaras can be a painstaking process. But people - fear not, don't lose heart! If it was that easy - then where is the challenge ?

Also, for consistency sake I think we should come up with a common legend. I can not too attached to mine - and so I can change.

Arun

[arunk]

My current challenge is: There are many occasions in that short alapana where Abhogi reveals itself so beautifully in a very short interval. I want to get those including the nuances that make it the Abhogi it is.

This is very true and yes that would hopefully be what we will learn.

[cmlover]

Finally I got to d/l using Firefox!
Based on the first 11 secs I also thought it was sriranjani. Having heard it in full I agree it is abhogi.

I assumed she will end the aalaapana at aadhaara shadjam. As I was tempted I checked with my software and found it to be 208Hz which is 5.5 kaTTai. Now as I checked Arun's tara shadjam I got 418Hz bang on! His madhyama D2 checks out at 350Hz. Hence Arun appears to be vindicated! But I am ashamed I would not have gotten them without using the software. My poor ears:(

Now not to divert your attention I was attempting to 'svarapuduthify' (let me use SP hereafter) the first 11 secs. It was challenging. Now I am getting (by cheating) m d s'd d, m,, (don't know). (Now I promise I did not look at Arun's interpretation while writing this
Would you guys also like to try?

[rajeshnat]

Arun
I wish you gave the answers after giving us(atleast me) more time. I just downloaded the question , immediately saw the answer too (bad school habits still persist). From next series let us have a cut off start time for answering say from 48 hours after posting the question. That way all of us will think thru. I have already seen the answer , but still will try to answer .

If possible can we have the next question where we can take an alApanai of Shri ranjani to subtlely show the defining dots of shriranjani(I hope I am not pushing you more here!!!). That comparison will solidify the understanding and take me to a different plane like you,vk,suji etal...!!

[arunk]

rajesh - this is precisely the reason why we had separate threads You were supposed to refrain from looking here until you had an answer AND posted it. That way people who want to discuss their answers can do so in the meantime. But I do not mind having a initial mandatory wait time so that people who may want to participate in the discussion from the beginning "don't miss out". I didnt think of that until now.

But actually yesterday I posted my answer only because I thought no one was being bold enough. And to keep others who were perhaps being a bit hesitant to jump into (on their own - even if they will do so "invisibly")

Well - we will also learn how to do this better as we go. Sorry for spoiling your party.

SrIranjani immediately? Ok but wouldnt that be confusing to many? Or would it just be easier? I thought we should go to the other pentatonics (the more obvious ones ).

I am open to any and all suggestions.

Arun

[Suji Ram]

Arun,
you are spot on "tara R" 1.01 sec. It goes as RSD

Suji - I wonder if we may be looking at slightly different timespots. Between 00:57-0:58 (itunes) to about 1:01, she sings

ta - da - ri - na - na - na - aa.. - ri.. - ri - ri

(To me, this seems like )

d - s' - r - g' - r' - r' - (d)g', - r', - s' - s'

Can you confirm?

I basically indicated the later last ri there, and you may have indicated earlier ones?

I was going further beyond -it goes RSD where she lands on D around 1:04 followed by violinist repeating RSD. Is that correct??
I assumed you spotted that R. Time points are tricky. I had noted on a paper in my commute so did not cross check after I so your answer.

You are right with the other R in the :58, :59. As I pointed out in my answers the tara octave can be spotted easily in that time period.

I wish I had done swaras in all octave too. Now I am lazy to work on it again.

And thanks for the symbolsused to denote gamakas etc. I am too bad at writing notations or even looking at books as I do not learn with notations much.

[ninjathegreat]

wow! the only thing I could indentify identify was the raga as Abhogi!!!!

[ninjathegreat]

G2 (tara): very briefly at 0:12 ta-na-na-ri

arun,

I don't seem to see this. There is a madhya G2, but the tara seems to occur at around 19s? Am I right?

[Suji Ram]

G2 (tara): very briefly at 0:12 ta-na-na-ri

arun,

I don't seem to see this. There is a madhya G2, but the tara seems to occur at around 19s? Am I right?

Both are madhya G
I did not hear tara octave until later beyong beyond 45 sec

Now analysing what others have is even more trickier.
I guess we should stick to certain time points in the question.

The fact that it is female voice madhya seems tara I guess for untrained ears

[arunk]

I was going further beyond -it goes RSD where she lands on D around 1:04 followed by violinist repeating RSD. Is that correct??

Could be - but I am not sure. I think this is where labeling gets tricky. Perhaps it is S'/R' S'/R',D' i.e the first ri is taken from sa (is sa anuswara or not? I have no idea). The second one is a slide that starts from sa to ri and comes back to da (via sa of course).

I think perhaps anuswara for sa here does not make sense as it is even more implicit (e.g. if ri is done with emphasis/nokku it would have a tinge of ga and thus ga is the anuswara?)

Others: Is this going too quickly too ahead? Do we need to construct simpler exercises like identify "quadrant" in one or two sections, identify swaras at specific spots where singer is holding it long enough?

Perhaps exercises should include both levels: simpler ones like above, then harder ones (like swara construction for a phrase)?

Also do you need more obvious proofs/demonstration as to why something is ri, da etc.? It is possible for some cases but will take effort.

[ninjathegreat]

ok, 20 second bit at the start. Just what I heard (I think!) (Question 4?)

0 10
ta da ra na aa ra na na na aa na aa ra na na ta da ra na na ra na na aa aa aa na aa aa ra

M D S' D ~ M M M MDS'DM D M G G M G G S R G D D ~ ~ ~ D S' G' S'

[cmlover]

The logic of determination explicitly (in simple terms) will be much more useful than the answers

[cmlover]

ninja
you are on the right track! Just explain how you went from ta da ri na...
to s r g m....

[ninjathegreat]

cml sir,

What I did was to listen to 10 second bits repeatedly till it was an infinite loop in my head - once the MDS'D (I think the clincher in identifying D was the gamakam at 3s) was identified in the beginning, it seemed to be easier to get the other swaras (heavy concentration for a really bad dabbler like me!!). Of course, I might be VERY wrong!! (untrained ear and faulty memory???)

To think that Lalgudi could do this effortlessly at age 2!!!

Ninja

[Suji Ram]

swara breakdown
Firstly I had to sing aro /ava for abhOgi a few times to get the swara Sruti.
I arrived at the first phrase from the familarity in the cittaswara 2 of abhogi varnam which goes as MDS /~DM
Once this was done my ears picked the Sruti of swaras following it. Some places gamakas are characteristic for the raga. I listen to swaras before and after.
Overall a tough task. Like everyone the familarity of phrases without breakdown is easier.

Arun,
We seem to have the same swara break down (except the way I have written down). Any experts apart from you who can confirm

[ramakriya]

I don't know if I can explain the logic of getting to the swaras - Most often the swaras just reveal themselves when I hear - I know this is not a very helpful answer for the purpose of this thread.
But it may be a question of exposure.

I know many people can repeat a phrase (mostly vocal) without being able to get to the swaras as such. Some instrumentalists (I am not talking about classically trained instrumentalists here - because if they look at the fret board, they should be able to tell it ) do it too. While I may not have the ability to reproduce each and every sangati exacly, I can do the dissection part (getting to the swaras) quite effectively! But I really wish I had the former than the latter

-Ramakriya

[cmlover]

Let us keep in mind that she starts singing only after the 3rd second. Hence our time coordinates should refer to time zero to be consistent!

[Suji Ram]

While I may not have the ability to reproduce each and every sangati exacly, I can do the dissection part (getting to the swaras) quite effectively! But I really wish I had the former than the latter

-Ramakriya

Can you comment on my post 17. I thought I heard upper octave not until 53 secs. May be an occasional R` but not G`

[cmlover]

This discussion is quite helpful.
Repeating the aro/avaro in the three speeds is the first step.
Familiarity with the raga phrases is a prerequisite. This can be got by listening to a number of renditions. Especially the kalpana svaram singing by experts heard repeatedly will familiarize with the common phrases.
Now one should practise the phrases by humming or ta da ri na naa quite a few times to internalize them.
Once done while listening to a new piece one should be able to map the known phrases in place. Of course there will be gaps due to unknown phrases (or even mistakes by the performer

Am I right so far?

[arunk]

We seem to have the same swara break down (except the way I have written down).

Yes we do. I checked:

Code: Select all

Suji: M D S ~D  , M M M mD M,  R  ~G  M ,, G R R
Arun: M D S' D~ , M M M D  M (r) G~ (r)/M G R R ,

Ninja and us do for while (he has included the sa tinge with da - when da is done with emphasis). but they deviate considerably after a bit.

Code: Select all

Suji : M D S ~D  , M M M mD M,    R ~G   M ,, G R R
Arun : M D S' D~ , M M M D  M   (r) G~ (r)/M G R R ,
Ninja: M D S' D~   M M M MD S'DM  D  M   G  G  M  G G  S

As you can see it deviates from D M G part (i included only up to 10s marker i.e. first phrase).

[Suji Ram]

Thanks Arun,
while playing abhogi on violin I play D from M that's the reason I included mD M. It is kind of a slur.

[arunk]

My process:

1. First is to get the raga - it sort of clicks as I listen.
2. Once it does, even though I still cannot tell swaras, I can follow the contour - and I think I almost always extrapolate on my own all the way to madya "sa".
3. Once I establish the sruthi and raga (and thus its structure) there is a probably an internalization of the raga (in the sruthi of the sample). After that I think I have reasonable hande of the swarasthanas (mainly from my lessons).
4. Then to or identifying individual swaras - say the prominent ones that are held flat, I can usually sing the swarasthana and match it. Sometimes it takes a couple of attempts and there are misfires. Certain combinations can be hard if they are quick R2G2, D2N2 G3M1 and neither is held prominently or shown with gamaka.

For the non-flat ones, one must have some handle/idea of the possible gamakas for various swaras of the raga. Some are perhaps obvious even if one has not learnt the raga formally but has a reasonable idea of swarasthanam - as they are very prominent and characteristic of raga (say da here for Abhogi). Some you can correlate against other ragas.

In general, a good idea of the swarasthanams is absolutely essential to make any initial progress. With this you can perhaps even decipher the swaras oscillated in a kampita gamaka. To get the "melodic feel" of combinations of swaras for the raga would help - e.g. how does MDS'D~ sound, S'D~ etc. sound. For this listening to kalpanaswara (slow speed) can help.

[arunk]

Thanks Arun,
while playing abhogi on violin I play D from M that's the reason I included mD M. It is kind of a slur.

IMO, your representation is more valid ( i need to check - assuming it is so, I should have indicated /D in mine)

[arunk]

I checked - it is indeed D from M.

Ninja's indication Sa I think also applies perhaps as an anuswara. In Abhogi, da is commonly rendered assertively (nokku?) and will have a tinge of sa. A very good, prominent example of this is at the start of 35s - the same kind of da that figures in "bA" of marubA.....ri in the caraNam of the evaribOdana varnam

[Suji Ram]

learning process for swara breakdown exercise (additional)
1. learning varnams in swaras in different ragas (covering all 12 swarasthanas)helps a lot in getting the swarasthanams right. I feel it is a MUST.

2. Phrases can be learnt in varnams as well as kritis. I prefer kritis because the pauses and gamakas are brought out nicely by composers and then there is variety.

3. Constant humming or dabbling of phrases.

[Suji Ram]

I checked - it is indeed D from M.

Ninja's indication Sa I think also applies perhaps as an anuswara. In Abhogi, da is commonly rendered assertively (nokku?) and will have a tinge of sa. A very good, prominent example of this is at the start of 35s - the same kind of da that figures in "bA" of marubA.....ri in the caraNam of the evaribOdana varnam

Gamaka for D: ds,ds,d It does have jaru type gamaka

I indicated in my post but I did not include for simplicity

[ramakriya]

While I may not have the ability to reproduce each and every sangati exacly, I can do the dissection part (getting to the swaras) quite effectively! But I really wish I had the former than the latter

-Ramakriya

Can you comment on my post 17. I thought I heard upper octave not until 53 secs. May be an occasional R` but not G`

What you say is true - at 51 sec there is a R' (which might have a slight anusvara from G when you play on an instrument) - But a straigt G' is heard only at 58 sec.

-Ramakriya

[arunk]

Only now I realize that maybe a typo in my post caused this tara stayi discussion (i was a bit puzzled by it )

I said in my answer:

G2 (tara): very briefly at 0:12 ta-na-na-ri . The ta-na-na-ri is d-g'-d'-/s (da, tara-ga, slide down to back to da, and slide up to tara-sa)

I meant 1:12!

Oops - Sorry! Tara stayi does not figure at 0:12 - only later.

Arun

[cmlover]

suji
very good points!
Since varnams are authoritative, practising them will give a good grip on the 'key' phrases. I also try singing the whole varnam using 'akaara'. That strengthens the svarasthanas since there will be no consonants to interfere! Of course I record it an hear to make sure that the raga bhava is fully preserved. At this stage one should not expect 100% accuracy since even experts do not have that kind of accuracy
As I practice the 'akaara' I consciously try to observe where the sound emanates in my 'physiological system' (a la shobillu ). Since I know the notes I can confirm the note identification with the 'akaaram' produced.
Now for the unknown phrase I try to convert it into 'akaara' and hum it. Due to the association I can guess the notes of the phrase at least approximately. As I have been trying this quite recently I am getting close to 80% succcess rate. The problem is that I do not have the 'correct answer' to the unknown phrases that I am trying to identify to verify my accuracy. It will be nice to develop a 'bank' of aalaapana phrases with solutions to test oneself! Are we game?

[ramakriya]

Only now I realize that maybe a typo in my post caused this tara stayi discussion (i was a bit puzzled by it )

I said in my answer:

G2 (tara): very briefly at 0:12 ta-na-na-ri . The ta-na-na-ri is d-g'-d'-/s (da, tara-ga, slide down to back to da, and slide up to tara-sa)

I meant 1:12!

Oops - Sorry! Tara stayi does not figure at 0:12 - only later.

Arun

@1:12 Sheela sings : da ri da sa

[arunk]

Gamaka for D: ds,ds,d It does have jaru type gamaka

I indicated in my post but I did not include for simplicity

I get confused by the terminology (as it seems a lot of them seemed to be from a vINa angle.

So avoding the term slide, jAru, kampita: IMO, this da is like a ascent from ma with force to touch sa briefly and then descent da - all as a single note. It may then be followed by oscillation between da and sa. Would you agree?

[arunk]

Only now I realize that maybe a typo in my post caused this tara stayi discussion (i was a bit puzzled by it )

I said in my answer:

G2 (tara): very briefly at 0:12 ta-na-na-ri . The ta-na-na-ri is d-g'-d'-/s (da, tara-ga, slide down to back to da, and slide up to tara-sa)

I meant 1:12!

Oops - Sorry! Tara stayi does not figure at 0:12 - only later.

Arun

@1:12 Sheela sings : da ri da sa

Very possible. I was flip-flopping between ga and ri there for a while

[arunk]

of course once someone else points it out (as if to break a tie), when i hear it now, it seems unambiguously ri - lol !

[Suji Ram]

Suji Ram]Gamaka for D: ds,ds,d
IMO, this da is like a ascent from ma with force to touch sa briefly and then descent da - all as a single note. It may then be followed by oscillation between da and sa. Would you agree?

Correct.
The descent D from S is slightly different.

Thanks Ramakriya

[arunk]

Correct.
The descent D from S is slightly different.

Agreed - it is slower - caressing. You get up to sa with force and then you "let it float down" to da. (hey - wait a minute, this sounds like tai chi ) ?

[Suji Ram]

Agreed - it is slower - caressing. You get up to sa with force and then you "let it float down" to da. (hey - wait a minute, this sounds like tai chi )

great imagery

[Suji Ram]

What about mandara M at 2:47, I can hear violin morphing voice perhaps,
and tara M- 1:57 Is that correct?

[arunk]

I think you are correct on both counts

tara M - also at 1:38 (and violin touches at 1:39-1:40)?

cmlover - if it helps, for these, I usually try have an idea as to what the pattern is and sing the swaras, and see if it matches. For example, here I try m,gr. If in case, ma I am sing sounded higher (it didnt), then I would have tried g-,rs

[cmlover]

Arun
What one needs is an iron-clad shruti (which unfortunately I don't have) to hit the right notes. I will not worry too much about the gamaka at early stages since they can be slurred (vocally).

I will love some guidance on vocally developing the 'nOkku' since it is very fleeting.

I find the quadrant idea of VK very helpful! (why didn't I think of that! Thanks VK for the idea!

[Suji Ram]

I think you are correct on both counts

tara M - also at 1:38 (and violin touches at 1:39-1:40)?

Ya, 1:38 , 1:57 are the only 2 regions where higher octaves are clear.

[vasanthakokilam]

I find the quadrant idea of VK very helpful!

Patent pending

[cmlover]

I find the quadrant idea of VK very helpful!

Patent pending

Ha ha!
Let me grab a 'trident' before you sue (attack