YouTube2Text

What You Need to Know About Suno's AI Music Detector — Transcript

by Music Tech Info · 1,081 words · 171 segments · language en · Watch on YouTube

Full transcript

  1. 0:01Okay, so this is something I wasn't
  2. 0:02expecting. This is called Sunno
  3. 0:04credentials. And basically what Sunno
  4. 0:05are doing is they've built an A AI music
  5. 0:09detector specifically for Sunno tracks.
  6. 0:13They've explained how important music is
  7. 0:16and that their latest wave of technology
  8. 0:19has made it so good for people to find
  9. 0:21and listen to music. And in terms of
  10. 0:24being responsible for the next wave,
  11. 0:27their mission is to build the world's
  12. 0:30best platform for music creation and to
  13. 0:32have a healthy and growing ecosystem.
  14. 0:35They've set a clear set of principles
  15. 0:39originally that has always been rooted
  16. 0:41in what they do. And in line with that
  17. 0:43and responsible use of Gen AI, they're
  18. 0:46starting to roll out new transparency
  19. 0:48tools that align with emerging industry
  20. 0:50standards so that songs generated on
  21. 0:52Sunno can be identified if they are
  22. 0:54shared on other platforms. Now there's
  23. 0:58other laws that have come into effect
  24. 1:00around the world including the EU which
  25. 1:01I've recently reported on in one in as
  26. 1:03my digital twin in short form. uh
  27. 1:06platforms with Gen AI tools are being
  28. 1:08required to disclose when content has
  29. 1:10been generated with AI and many
  30. 1:12platforms are choosing to meet those
  31. 1:14obligations using the industry standard
  32. 1:16technology called C2PA
  33. 1:18which is an individual invisible set of
  34. 1:21metadata excuse me attached to a piece
  35. 1:23of content that can help distribution
  36. 1:25platforms determine whether AI was used
  37. 1:27to create it
  38. 1:29and if they choose to use that
  39. 1:31information.
  40. 1:33So there's a lot of important
  41. 1:34conversations happening around AI
  42. 1:36transparency right now and they want to
  43. 1:38provide more context about how the tools
  44. 1:40actually work. The changes will only
  45. 1:42apply to new songs that are downloaded
  46. 1:44on Sunno going forward. So they've built
  47. 1:47a tracker which you can anybody can
  48. 1:49access and basically it's only going to
  49. 1:52apply to those new songs not older
  50. 1:54generations of Sunno going forward. So
  51. 1:57anything that has already been
  52. 1:58downloaded from Sunno won't be detected
  53. 2:00on this new tool. They're doing this
  54. 2:03because they want to improve
  55. 2:04transparency, but keeping artists
  56. 2:06creative freedom and the way they choose
  57. 2:08to use AI in their creative workflows.
  58. 2:11They believe that artists and platforms
  59. 2:13should decide how they want to discuss
  60. 2:15the use of AI tools with their
  61. 2:16audiences. And their role is to provide
  62. 2:19transparency tools. So, they've got more
  63. 2:21information on that for anybody that's
  64. 2:23interested if you go to
  65. 2:24suno.com/suno-credentials.
  66. 2:27So, these content credentials are
  67. 2:29designed to help platforms meet their
  68. 2:31own transparency obligations, but they
  69. 2:33are not a judgment about the value,
  70. 2:35authenticity, or ownership of an
  71. 2:37artist's work and expression. These
  72. 2:40tools also don't change how a song
  73. 2:42sounds or affect ownership. Instead,
  74. 2:44they simply provide information that
  75. 2:46platforms can choose to use in their own
  76. 2:48discretion.
  77. 2:50So actually on this page you're able to
  78. 2:52use the tool by adding in a URL
  79. 2:56or an audio file and it's up to 100
  80. 3:00megabytes in terms of that. Now there
  81. 3:03are some FAQs which I'm just going to
  82. 3:04quickly go through sooner built this
  83. 3:07because of their transparency. Obviously
  84. 3:09EU regulations and other regul
  85. 3:11regulations around the world are kind of
  86. 3:15leaning towards this now uh because AI
  87. 3:18platforms for the EU now have to say
  88. 3:20whether they've used AI tools or not. Um
  89. 3:23it's based off CTPA which is an open
  90. 3:26industry standard used across the tech
  91. 3:28industry that shows where a piece of
  92. 3:30information sorry content came from and
  93. 3:32how it was actually made. When a song is
  94. 3:34downloaded from Sunno, we attach
  95. 3:36metadata identifying it as AI generated
  96. 3:38content made with Sunno.
  97. 3:41This allows anyone to check whether a
  98. 3:44song carries Sunno's content
  99. 3:46credentials. You can upload an audio
  100. 3:49file or paste a public link and we'll
  101. 3:51tell you whether we find credentials
  102. 3:52showing it was made by Sunno or not. Can
  103. 3:55other tools read Sunno's content
  104. 3:58credentials? Yes, they can because this
  105. 4:00tool is using C2PA which is a standard
  106. 4:03same open standard used by other content
  107. 4:07companies which means you don't have to
  108. 4:09specifically use Sunno's tool to check a
  109. 4:11song is made with Sunno or not but
  110. 4:14they've made this available anyway and
  111. 4:17any Apple service that supports content
  112. 4:19credentials can read the label in a
  113. 4:21sunno file and the label travels where
  114. 4:23the file wherever it's shared and there
  115. 4:27is an API as well for anybody body
  116. 4:28that's interested in that. So, this is
  117. 4:31quite interesting. Now, let me have a
  118. 4:33look. I'm just going to pause and drop
  119. 4:35in an audio file from a while ago. So,
  120. 4:37I've just dropped in this little clip of
  121. 4:39a song. Uh, this was probably 2 years
  122. 4:41ago. It's not made by So, checking
  123. 4:46nose content credentials found that
  124. 4:48happened very quickly. There is a
  125. 4:50feedback button as well. So, let's try
  126. 4:53another one. Now, this is one that was
  127. 4:56part generated by Sunno. And you can see
  128. 4:59I mastered it as well. So, let's check
  129. 5:01that one. This was a while ago, by the
  130. 5:03way, going back to around October last
  131. 5:05year.
  132. 5:08No Sunseo content credentials found.
  133. 5:11Okay, let's check another one.
  134. 5:14Now, this song is by Sunno. I haven't
  135. 5:17messed with it at all. And literally, I
  136. 5:20downloaded it on September the 8th. So,
  137. 5:24this
  138. 5:25launched. It doesn't actually give a
  139. 5:27date, does it? But I believe it was
  140. 5:28September the 8th. So, let's check this
  141. 5:30one. This should be a positive.
  142. 5:34Oh, interesting. Okay. Maybe that song
  143. 5:37was downloaded before this has happened.
  144. 5:42Okay. So, now I've literally just gone
  145. 5:44in to Sunno and I've just generated a
  146. 5:47song, a test song. It's called Warm
  147. 5:49Groove. Uh this was after this was
  148. 5:51implemented on September the 9th 1250.
  149. 5:55So now let's check.
  150. 5:59There we go. So content credentials
  151. 6:02found it's giving the content ID I
  152. 6:06presume of the track exactly when it was
  153. 6:09created. Um, so it's 2 minutes ago
  154. 6:12actually. Uh, by Sunno Incorporated,
  155. 6:16system name Sunno
  156. 6:18and even the model number.
  157. 6:22So there you have it. I think that's
  158. 6:23quite interesting to be fair. It's
  159. 6:25interesting that they haven't included
  160. 6:27past generations.
  161. 6:29Um, but they're doing it from now. So
  162. 6:32any generation in a nutshell that you do
  163. 6:35from now will be picked up with this
  164. 6:37detector. Don't forget there are APIs to
  165. 6:41allow different websites and platforms
  166. 6:42to use it. And yeah, this is where we
  167. 6:46are.
  168. 6:48Let me know what you think in the
  169. 6:49comments
  170. 6:51and I'll catch you on the next one. Take
  171. 6:52care. bite.

About this transcript

This page contains the full transcript of What You Need to Know About Suno's AI Music Detector by Music Tech Info, generated from the public captions YouTube serves with the video. The transcript has 1,081 words across 171 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.