Add audio samples and spectrogram comparison to README.md; include original and reconstructed audio files

Files changed (6) hide show

.gitattributes CHANGED Viewed

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text

README.md CHANGED Viewed

@@ -15,6 +15,19 @@ tags:
 A fine-tuned variant of [Qwen/Qwen3-TTS-Tokenizer-12Hz](https://huggingface.co/Qwen/Qwen3-TTS-Tokenizer-12Hz) that decodes speech tokens to **48 kHz** audio instead of the original 24 kHz — with no custom code required.
 ## Overview
 The original Qwen3-TTS-Tokenizer-12Hz decoder outputs audio at 24 kHz. This model extends it by appending one additional `DecoderBlock` (upsample rate ×2) to the existing decoder stack, effectively doubling the output sample rate to 48 kHz.

 A fine-tuned variant of [Qwen/Qwen3-TTS-Tokenizer-12Hz](https://huggingface.co/Qwen/Qwen3-TTS-Tokenizer-12Hz) that decodes speech tokens to **48 kHz** audio instead of the original 24 kHz — with no custom code required.
+## Audio Samples
+Sample audio encoded with `Qwen/Qwen3-TTS-Tokenizer-12Hz` and decoded with both the original 24 kHz model and this 48 kHz model.
+| | Audio |
+|---|---|
+| Original<br/>[Voice material by “Amitaro’s Voice Material Studio”](https://amitaro.net/) | <audio controls><source src="assets/original.wav" type="audio/wav"></audio> |
+| Reconstructed 24 kHz (base model) | <audio controls><source src="assets/reconstructed_24k.wav" type="audio/wav"></audio> |
+| Reconstructed 48 kHz (this model) | <audio controls><source src="assets/reconstructed_48k.wav" type="audio/wav"></audio> |
+![Mel spectrogram comparison](assets/spectrogram_comparison.png)
 ## Overview
 The original Qwen3-TTS-Tokenizer-12Hz decoder outputs audio at 24 kHz. This model extends it by appending one additional `DecoderBlock` (upsample rate ×2) to the existing decoder stack, effectively doubling the output sample rate to 48 kHz.

assets/original.wav ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:7a5475679d1a68109c5abea2ab72f172856f3f105880dd9ccbe402e9334bf4d0
+size 413856

assets/reconstructed_24k.wav ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:adb8d5dbe43d69c9a8634c8c03f38ae6c1662588d2b7600f8b99ff5f25613acb
+size 207404

assets/reconstructed_48k.wav ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:9a9d2055880b76e452dcfcee4634b3d7c40d18d493d87b9c759ca4809e8c190c
+size 414764

assets/spectrogram_comparison.png ADDED Viewed