# Utterance-level classification on ARCH (RAVDESS, EMOVO, AudioMNIST, SLURP) post-VQ 57 Source: [Utterance-level classification on ARCH (RAVDESS, EMOVO, AudioMNIST, SLURP) post-VQ 57]() Collection index: [llms.txt]() Data updated: 2026-06-30T08:27:03.349Z Evaluation of utterance-level classification performance across the combined ARCH suite, specifically using the RAVDESS, EMOVO, AudioMNIST, and SLURP components processed via the post-VQ 57 configuration. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Utterance-level classification | | Dataset (including any stated split) | ARCH (RAVDESS, EMOVO, AudioMNIST, SLURP) post-VQ 57 | | Metric | RAVDESS Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [RAVDESS Accuracy](): Higher is better; unit: PERCENT. - [RAVDESS F1 Score](): Higher is better; unit: PERCENT. - [EMOVO Accuracy](): Higher is better; unit: PERCENT. - [EMOVO F1 Score](): Higher is better; unit: PERCENT. - [AudioMNIST Accuracy](): Higher is better; unit: PERCENT. - [AudioMNIST F1 Score](): Higher is better; unit: PERCENT. - [SLURP Accuracy](): Higher is better; unit: PERCENT. - [SLURP F1 Score](): Higher is better; unit: SCALAR. - [ARCH Accuracy](): Higher is better; unit: PERCENT. - [ARCH F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | RAVDESS Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DTM-Codec@50Hz | 37.85 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@40Hz | 37.5 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@80Hz | 37.5 | Not recorded | 6a434d7dba73165dbb743978 | | X-Codec 2.0 | 37.15 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@80Hz | 36.81 | Not recorded | 6a434d7dba73165dbb743978 | | BigCodec | 36.11 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@25Hz | 35.42 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@50Hz | 34.72 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@40Hz | 32.99 | Not recorded | 6a434d7dba73165dbb743978 | | DTM-Codec@25Hz | 32.64 | Not recorded | 6a434d7dba73165dbb743978 | | VARSTok | 27.43 | Not recorded | 6a434d7dba73165dbb743978 | | WavTokenizer | 27.43 | Not recorded | 6a434d7dba73165dbb743978 | | VARSTok | 27.08 | Not recorded | 6a434d7dba73165dbb743978 | | WavTokenizer | 24.65 | Not recorded | 6a434d7dba73165dbb743978 | | VARSTok | 24.31 | Not recorded | 6a434d7dba73165dbb743978 | ### DTM-Codec@50Hz RAVDESS Accuracy: 37.85. Result ID: 6a437dbcfbc373592c948cbd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"VFR","\|C\|":"16,384"}. This record does not assert independent reproduction. ### DTM-Codec@40Hz RAVDESS Accuracy: 37.5. Result ID: 6a437dbcfbc373592c948cbc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"FFR","\|C\|":"65,536"}. This record does not assert independent reproduction. ### DTM-Codec@80Hz RAVDESS Accuracy: 37.5. Result ID: 6a437dbcfbc373592c948cbf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"VFR","\|C\|":"16,384"}. This record does not assert independent reproduction. ### X-Codec 2.0 RAVDESS Accuracy: 37.15. Result ID: 6a437dbcfbc373592c948cc1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"50Hz","\|C\|":"65,536","Features":"Wav2Vec2-BERT semantic features"}. This record does not assert independent reproduction. ### DTM-Codec@80Hz RAVDESS Accuracy: 36.81. Result ID: 6a437dbcfbc373592c948cc0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"FFR","\|C\|":"65,536"}. This record does not assert independent reproduction. ### BigCodec RAVDESS Accuracy: 36.11. Result ID: 6a437dbcfbc373592c948cc2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"80Hz","\|C\|":"8,192"}. This record does not assert independent reproduction. ### DTM-Codec@25Hz RAVDESS Accuracy: 35.42. Result ID: 6a437dbcfbc373592c948cba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"FFR","\|C\|":"65,536"}. This record does not assert independent reproduction. ### DTM-Codec@50Hz RAVDESS Accuracy: 34.72. Result ID: 6a437dbcfbc373592c948cbe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"FFR","\|C\|":"65,536"}. This record does not assert independent reproduction. ### DTM-Codec@40Hz RAVDESS Accuracy: 32.99. Result ID: 6a437dbcfbc373592c948cbb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"VFR","\|C\|":"16,384"}. This record does not assert independent reproduction. ### DTM-Codec@25Hz RAVDESS Accuracy: 32.64. Result ID: 6a437dbcfbc373592c948cb9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"VFR","\|C\|":"16,384"}. This record does not assert independent reproduction. ### VARSTok RAVDESS Accuracy: 27.43. Result ID: 6a437dbcfbc373592c948cc3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"τ=0.8","\|C\|":"4,096"}. This record does not assert independent reproduction. ### WavTokenizer RAVDESS Accuracy: 27.43. Result ID: 6a437dbcfbc373592c948cc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"75Hz","\|C\|":"4,096"}. This record does not assert independent reproduction. ### VARSTok RAVDESS Accuracy: 27.08. Result ID: 6a437dbcfbc373592c948cc4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"τ=0.7","\|C\|":"4,096"}. This record does not assert independent reproduction. ### WavTokenizer RAVDESS Accuracy: 24.65. Result ID: 6a437dbcfbc373592c948cc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"40Hz","\|C\|":"4,096"}. This record does not assert independent reproduction. ### VARSTok RAVDESS Accuracy: 24.31. Result ID: 6a437dbcfbc373592c948cc5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"τ=0.6","\|C\|":"4,096"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.