Meta: Speech Metrics and Feature Extraction Pipelines

[satra]

Umbrella issue for adding new speech feature extraction capabilities.

Sub-issues

• Task: Add models/pipelines to extract Quantity of speech metrics #154 Add models/pipelines to extract Quantity of speech metrics
• Task: Add models/pipelines for extracting Rate of speech metrics #155 Add models/pipelines for extracting Rate of speech metrics
• Task: Add models/pipelines for extracting Volume (tone of Speech) metrics #156 Add models/pipelines for extracting Volume (tone of Speech) metrics
• Task: Add models/pipelines to extract Fluency & Rhythm metrics #157 Add models/pipelines to extract Fluency & Rhythm metrics
• Task: Add models/pipelines for extracting Response latency metrics #158 Add models/pipelines for extracting Response latency metrics
• Task: Add models/pipelines for extracting "Other aspects of speech" metrics #159 Add models/pipelines for extracting "Other aspects of speech" metrics
• Task: Add models/pipelines to extract auditory Affect metrics #160 Add models/pipelines to extract auditory Affect metrics
• [Task] Extend the list of supported audio features #193 Extend the list of supported audio features
• Task: Add Creak Feature #346 Add Creak Feature
• Task: Add Hear Embedding Extraction #364 Add Hear Embedding Extraction

Scope

Expand senselab's speech analysis capabilities with new metrics for quantity, rate, volume, fluency, rhythm, response latency, affect, creak detection, and HEAR embeddings. Many of these can share infrastructure (feature extraction pipeline, model loading patterns).