Last active
March 6, 2024 17:13
-
-
Save EnisBerk/d49edb61e97cca03482a3644e358ed0a to your computer and use it in GitHub Desktop.
prompts for the inference experiments
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# source for Grouse description: https://birdsoftheworld.org/bow/species/wilpta/cur/introduction#vocal | |
{ | |
'default': | |
'write an audio caption describing the sound', | |
'label_in_prompt': | |
'', | |
'grouse01': | |
"""provide labels for the audio file, could it be a Grouse? | |
here is detailed information on Grouse call types: | |
Both Sexes. (1) Kok is a short (50 ms) clucking call; \ | |
volume and frequency vary with intensity of arousal\ | |
(2) Ko-ko-ko is a low-amplitude call given \ | |
in prolonged bouts, sounding like low growls; \ | |
(3) Krrow is a medium-length (50-300 ms)\ | |
call that rises quickly and falls slowly in frequency: In males, sounds \ | |
like bugow; in females, like meow; given during aggressive disputes \ | |
(4) Kohwa, Kohway, and Kohwayo are medium-length (100 ms) \ | |
calls often given in association with \ | |
Krrow during aggressive interactions. \ | |
(5) Aroo is a variable call with falling \ | |
and rising pattern of frequency modulation at varying rate and intensity. (6) \ | |
Rattle (song on the ground) is a long (800 ms), accelerating string of short \ | |
elements similar to Kok\ | |
(7) Flight Song (Aerial Bek) is a 2-part vocalization; first part is a decelerating \ | |
series of modified Ko-ko-ko calls, second part a decelerating Kohwa alternatively \ | |
described as a series of nasal barks, typically a few single notes followed by a rattling \ | |
sequence of 8–12 guttural notes, and several doubled notes on landing, “whek!..whek-kekekrrrrekek-kek...\ | |
koh-wa..koh-wa..koh-wa". (8) Scream is a brief, high-frequency\ | |
call with indeterminate harmonic structure. (9) Hiss is a band of white noise about 2 s long.\ | |
list labels for the audio file: | |
""", | |
'grouse02_label_in_prompt': | |
"""provide labels for the audio file, could it be a Grouse? | |
here is detailed information on Grouse call types: | |
Both Sexes. (1) Kok is a short (50 ms) clucking call; \ | |
volume and frequency vary with intensity of arousal\ | |
(2) Ko-ko-ko is a low-amplitude call given \ | |
in prolonged bouts, sounding like low growls; \ | |
(3) Krrow is a medium-length (50-300 ms)\ | |
call that rises quickly and falls slowly in frequency: In males, sounds \ | |
like bugow; in females, like meow; given during aggressive disputes \ | |
(4) Kohwa, Kohway, and Kohwayo are medium-length (100 ms) \ | |
calls often given in association with \ | |
Krrow during aggressive interactions. \ | |
(5) Aroo is a variable call with falling \ | |
and rising pattern of frequency modulation at varying rate and intensity. (6) \ | |
Rattle (song on the ground) is a long (800 ms), accelerating string of short \ | |
elements similar to Kok\ | |
(7) Flight Song (Aerial Bek) is a 2-part vocalization; first part is a decelerating \ | |
series of modified Ko-ko-ko calls, second part a decelerating Kohwa alternatively \ | |
described as a series of nasal barks, typically a few single notes followed by a rattling \ | |
sequence of 8–12 guttural notes, and several doubled notes on landing, “whek!..whek-kekekrrrrekek-kek...\ | |
koh-wa..koh-wa..koh-wa". (8) Scream is a brief, high-frequency\ | |
call with indeterminate harmonic structure. (9) Hiss is a band of white noise about 2 s long.\ | |
Write an audio caption describing the sound. \ | |
Could the sound be Silence, Biological soundscape,\ | |
aviation noise, Rainfall, Grouse, Insect,\ | |
Songbird, Duck and/or Goose and/or Swan,\ | |
anthropogenic noise, physical soundscape, Bird, or Wind?' | |
""" | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment