May 282020

Una Chow


  12 Responses to “Voice Quality of Gitksan Ejectives”

  1. Hi Una! Interesting work! Have you considered looking at F2′ (F2 prime, or “effective F2”; Carlson et al. 1970, Carlson et al. 1975)? Since you’re finding a strong effect with both F2 and F3, it seems like a combination of them might also give you an effect, and F2′ is a reasonable place to look for that.

    • Hi Nathan,

      Sorry for the delay in responding as I did not see the comment here during the poster session and I had to leave shortly after the session.

      Thank you for reading my poster and for your helpful suggestion.

      I am curious to look at F2′. I have not had any experience with it. If I understand it correctly, its calculation involves all first four formants. Perhaps I should also explore H1-A4 for comparison or a modified F2′ without F4.


      • Thanks Una!

        F2′ is usually conceived of as an integration of formants above F1, so since it excludes F1, it seems potentially useful here, as the formants above F1 seem to be the ones that carry the most information about stop type.

        That said, F2′ is not a precisely defined concept, so it’s a bit tricky to use. Whose particular formula do you use, etc. And since F2′ is a perceptual measure, rather than a raw, purely acoustic measure, it may not even be appropriate to mix it in here.

        I guess it depends on what the goal is. Are you trying to find a way for phoneticians to distinguish stop types in Gikstan, or are you trying to figure out what Gikstan speakers are latching on to when they distinguish stop types? This could lead you to explore very different measures!

        • Hi Nathan,

          Thanks for the explanation. At this point, I am exploring the acoustic aspect of these stops. It’s more in the direction of the first part of your question, although I am interested in inter-speaker variation in producing these stops as well.

          F2′ may be appropriate if I get to the perceptual analysis so it’s good to learn about it. I found Carlson et al. (1975) in Fant’s book so I’ll have a look at that.


  2. Hi! This is interesting! Overall, the acoustic measures you looked at seem to give more robust cues to stop type for the velars and uvulars than for the alveolars and palatals. Do you have any sense that other factors might play a more important role in the alveolar and palatal ejectives?

    • Hi Sara,

      Sorry for the delay in responding as I did not see the comment here during the poster session and I had to leave shortly after the session.

      Thank you for reading my poster and for bringing to my attention the cueing effect on velars/uvulars versus alveolars/palatals.

      I looked at the harmonic-to-noise ratio as well. The statistical test results suggest that it is a robust cue to stop type for alveolars/palatals but not for velars/uvulars. However, it applies to speaker HH only. No significant difference was found for the other speaker.

      Also, cepstral peak prominence can serve as a robust cue for all four places of articulation (based on statistical results), but again, for HH only.

      I would be interested in exploring F2′, as Nathan suggests, to see if H1-A2′ would make a stronger cue for all four places of articulation for both speakers. It involves F4, which I have not looked at.


  3. Like!! Really appreciate you sharing this blog post.Really thank you! Keep writing.

  4. I always spent my half an hour to read this web site’s articles or reviews daily along with a mug of coffee.

  5. Thanks so much for the blog post.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>