# Discrete Audio Representation as an Alternative to Mel-Spectrograms for Speaker and Speech Recognition

arXiv: https://arxiv.org/abs/2309.10922 · HTML: https://mlsanity.com/paper/2309.10922/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Sep 2023.

## Built on by (most cited first, 11 of 11)

- [BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data](https://mlsanity.com/paper/2402.08093/) (2024)
- [Autoregressive Speech Synthesis without Vector Quantization](https://mlsanity.com/paper/2407.08551/) (2024)
- [Recent Advances in Discrete Speech Tokens: A Review](https://mlsanity.com/paper/2502.06490/) (2025)
- [DASB - Discrete Audio and Speech Benchmark](https://mlsanity.com/paper/2406.14294/) (2024)
- [Codec-ASR: Training Performant Automatic Speech Recognition Systems with Discrete Speech Representations](https://mlsanity.com/paper/2407.03495/) (2024)
- [FELLE: Autoregressive Speech Synthesis with Token-Wise Coarse-to-Fine Flow Matching](https://mlsanity.com/paper/2502.11128/) (2025)
- [Speech Synthesis From Continuous Features Using Per-Token Latent Diffusion](https://mlsanity.com/paper/2410.16048/) (2024)
- [Towards Audio Codec-based Speech Separation](https://mlsanity.com/paper/2406.12434/) (2024)
- [StreamMel: Real-Time Zero-shot Text-to-Speech via Interleaved Continuous Autoregressive Modeling](https://mlsanity.com/paper/2506.12570/) (2025)
- [Codec2Vec: Self-Supervised Speech Representation Learning Using Neural Speech Codecs](https://mlsanity.com/paper/2511.16639/) (2025)
- [Explainable Disentanglement on Discrete Speech Representations for Noise-Robust ASR](https://mlsanity.com/paper/2510.25150/) (2025)

## Abstract (arXiv)

Discrete audio representation, aka audio tokenization, has seen renewed interest driven by its potential to facilitate the application of text language modeling approaches in audio domain. To this end, various compression and representation-learning based tokenization schemes have been proposed. However, there is limited investigation into the performance of compression-based audio tokens compared to well-established mel-spectrogram features across various speaker and speech related tasks. In this paper, we evaluate compression based audio tokens on three tasks: Speaker Verification, Diarization and (Multi-lingual) Speech Recognition. Our findings indicate that (i) the models trained on audio tokens perform competitively, on average within 1% of mel-spectrogram features for all the tasks considered, and do not surpass them yet. (ii) these models exhibit robustness for out-of-domain narrowband data, particularly in speaker tasks. (iii) audio tokens allow for compression to 20x compared to mel-spectrogram features with minimal loss of performance in speech and speaker related tasks, which is crucial for low bit-rate applications, and (iv) the examined Residual Vector Quantization (RVQ) based audio tokenizer exhibits a low-pass frequency response characteristic, offering a plausible explanation for the observed results, and providing insight for future tokenizer designs.