REVIEW 7 cited by
Deep Speaker: an End-to-End Neural Speaker Embedding System
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We present Deep Speaker, a neural speaker embedding system that maps utterances to a hypersphere where speaker similarity is measured by cosine similarity. The embeddings generated by Deep Speaker can be used for many tasks, including speaker identification, verification, and clustering. We experiment with ResCNN and GRU architectures to extract the acoustic features, then mean pool to produce utterance-level speaker embeddings, and train using triplet loss based on cosine similarity. Experiments on three distinct datasets suggest that Deep Speaker outperforms a DNN-based i-vector baseline. For example, Deep Speaker reduces the verification equal error rate by 50% (relatively) and improves the identification accuracy by 60% (relatively) on a text-independent dataset. We also present results that suggest adapting from a model trained with Mandarin can improve accuracy for English speaker recognition.
Forward citations
Cited by 7 Pith papers
-
Exploiting Neural Audio Codec Latents for Adversarial Audio Attacks
A conditional generator operating in neural audio codec latent space produces targeted adversarial audio examples in one forward pass, reaching up to 99% success rate at sub-7 ms inference.
-
The Hidden Cost of Pairwise Verification in Synthetic Speech Source Tracing
Global anchoring outperforms pairwise verification in synthetic speech source tracing by preserving more discriminative embedding directions, yielding lower error rates on in-domain and out-of-domain data.
-
Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations
A shared forward–inverse task representation uses auxiliary forward demos from novel configurations to execute inverse skills without inverse supervision, beating diffusion and multimodal VAE baselines.
-
An Exploration of ECAPA-TDNN and x-vector Speaker Representations in Zero-shot Multi-speaker TTS
In a fixed YourTTS framework, the H/ASP speaker encoder produces higher speaker similarity than x-vector and ECAPA-TDNN encoders.
-
Spatio-spectral diarization of meetings by combining TDOA-based segmentation and speaker embedding-based clustering
A cascade of TDOA-based spatial segmentation and speaker-embedding clustering diarizes meetings without multi-channel training data and handles overlapping speech and speaker position changes.
-
Towards Robust Uncertainty-Aware Speaker Modeling
Inter- and intra-speaker hardness in an uncertainty-aware softmax plus source-prior uncertainty calibration improves speaker verification reliability under domain shift.
-
Making deep neural networks work for medical audio: representation, compression and domain adaptation
A dissertation showing that transfer learning, tensor-compressed RNNs, and domain adaptation improve infant-cry models, while releasing the CryCeleb dataset for cry-based infant recognition.
Discussion (0). Continue with ORCID to comment.