AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Timers and Such: A Practical Benchmark for Spoken Language Understanding with Numbers

Loren Lugosch, Piyush Papreja, Mirco Ravanelli, Abdelwahab HEBA, Titouan Parcollet

NeurIPS 2021 · Large language models › Language, safety and interpretability · approach: Speech and audio language models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use the Timers and Such transcript language model instead of the LibriSpeech language modelASR-based SLU models · test-realaccuracy: 55.5% ± 3.4% -> 64.0% ± 3.3%Improved
Train the decoupled model with the Timers and Such transcript language model instead of the LibriSpeech language modelDecoupled · test-realaccuracy: 24.1% ± 1.1% -> 43.5% ± 2.0%Improved
Add synthetic training data to real training dataMultistage (TAS LM) · test-synthaccuracy: 51.5% ± 2.9% -> 85.4% ± 0.2%Improved
Train only on synthetic data rather than real plus synthetic dataDirect · test-realaccuracy: 77.5% ± 1.6% -> 68.0% ± 5.5%Hurt
Train directly from audio instead of using an ASR transcript as the intermediate inputSLU models trained on train-real · test-realaccuracy: 64.0% ± 3.3% -> 81.6% ± 5.4%Improved
Train directly from audio instead of using an ASR transcript as the intermediate inputSLU models trained on train-real + train-synth · test-synthaccuracy: 85.4% ± 0.2% -> 96.7% ± 0.3%Improved
Train the direct model on synthetic speech, which permits learning synthesizer-specific pronunciation patternsDirect · test-synthaccuracy: 70.0% ± 5.7% -> 96.7% ± 0.3%Improved
Use synthetic-only rather than real-only training dataDirect · test-realaccuracy: 81.6% ± 5.4% -> 68.0% ± 5.5%Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy