\n
![\"img\"](\"https://i.imgur.com/DsZvi8q.png\")
Here's a quick pictorial description of different kinds of spectrograms.
\nAll the images are generated using my 🐦BirdCLEF: Quick EDA with W&B kernel. Feedbacks appreciated. 😄✨🔥
\nThe frequency bins are linearly spaced in the y-axis.
\n
The frequency bins are spaced logarithmically. Note that by doing so lower frequency gets more say. Bird sounds are usually in the high-frequency range and thus we might not need this.
\n![\"img\"](\"https://i.imgur.com/XurLSZ4.png\")
![\"img\"](\"https://i.imgur.com/7k6JHJZ.png\")
![\"img\"](\"https://i.imgur.com/Y6EtlSP.png\")
![\"img\"](\"https://i.imgur.com/U7WcKch.png\")
![\"img\"](\"https://i.imgur.com/AyMagP9.png\")
The spectral centroid indicates at which frequency the energy of a spectrum is centered upon or in other words It indicates where the ” center of mass” for a sound is located.
\n![\"img\"](\"https://i.imgur.com/UQGlLn4.png\")
Note: I have not added Mel-Frequency Spectrogram. I will be updating it in the kernel as well as in this post. Previous audio-based competitions have massively used mel-scaled frequency spectrogram.
", "messages": [ { "id": 1266321, "postDate": "2021-04-07T16:57:46.313Z", "content": "Here's a quick pictorial description of different kinds of spectrograms.
\nAll the images are generated using my 🐦BirdCLEF: Quick EDA with W&B kernel. Feedbacks appreciated. 😄✨🔥
\nThe frequency bins are linearly spaced in the y-axis.
\n
The frequency bins are spaced logarithmically. Note that by doing so lower frequency gets more say. Bird sounds are usually in the high-frequency range and thus we might not need this.
\n
The spectral centroid indicates at which frequency the energy of a spectrum is centered upon or in other words It indicates where the ” center of mass” for a sound is located.
\n
Note: I have not added Mel-Frequency Spectrogram. I will be updating it in the kernel as well as in this post. Previous audio-based competitions have massively used mel-scaled frequency spectrogram.
", "rawMarkdown": "Here's a quick pictorial description of different kinds of spectrograms. \n\nAll the images are generated using my [🐦BirdCLEF: Quick EDA with W&B](https://www.kaggle.com/ayuraj/birdclef-quick-eda-with-w-b) kernel. Feedbacks appreciated. 😄✨🔥\n\n### Linear-Frequency Power Spectrogram\nThe frequency bins are linearly spaced in the y-axis. \n\n\n### Log-Frequency Power Spectrogram\nThe frequency bins are spaced logarithmically. Note that by doing so lower frequency gets more say. Bird sounds are usually in the high-frequency range and thus we might not need this. \n\n\n### Constant Q (Note) Power Spectrogram\n\n\n### Constant Q (Hz) Power Spectrogram\n\n\n### Tempogram with BPM markers\n\n\n### Chromagram\n\n\n## Additionally\n\n### Spectral Centroid\nThe spectral centroid indicates at which frequency the energy of a spectrum is centered upon or in other words It indicates where the ” center of mass” for a sound is located.\n\n\n**Note**: I have not added Mel-Frequency Spectrogram. I will be updating it in the kernel as well as in this post. Previous audio-based competitions have massively used mel-scaled frequency spectrogram. \n\n", "votes": 11 } ], "comments": [], "raw_markdown_by_id": { "1266321": "Here's a quick pictorial description of different kinds of spectrograms. \n\nAll the images are generated using my [🐦BirdCLEF: Quick EDA with W&B](https://www.kaggle.com/ayuraj/birdclef-quick-eda-with-w-b) kernel. Feedbacks appreciated. 😄✨🔥\n\n### Linear-Frequency Power Spectrogram\nThe frequency bins are linearly spaced in the y-axis. \n\n\n### Log-Frequency Power Spectrogram\nThe frequency bins are spaced logarithmically. Note that by doing so lower frequency gets more say. Bird sounds are usually in the high-frequency range and thus we might not need this. \n\n\n### Constant Q (Note) Power Spectrogram\n\n\n### Constant Q (Hz) Power Spectrogram\n\n\n### Tempogram with BPM markers\n\n\n### Chromagram\n\n\n## Additionally\n\n### Spectral Centroid\nThe spectral centroid indicates at which frequency the energy of a spectrum is centered upon or in other words It indicates where the ” center of mass” for a sound is located.\n\n\n**Note**: I have not added Mel-Frequency Spectrogram. I will be updating it in the kernel as well as in this post. Previous audio-based competitions have massively used mel-scaled frequency spectrogram. \n\n" } }