Speech understanding
What is speech understanding?
Speech understanding combines a transcript with acoustic cues that describe how the words were delivered — emotion, speaking style, prosody, and time-local change. It preserves more of the recorded signal than transcription alone: where speech recognition asks “what words were spoken?”, speech understanding asks “how were those words said?”oruk’s API covers a measured subset of this field — emotion, speaking style, and transcription of prerecorded files; see capabilities.
Speech understanding vs. speech recognition
Speech recognition (also called ASR or speech-to-text) converts audio into a string of words. It is a solved-enough commodity: many models transcribe accurately. But a transcript throws away tone, timing, and emphasis — exactly the signals that distinguish warm from sarcastic-sounding delivery, or a calm-sounding recorded caller from a frustrated one.
Speech understanding keeps those signals. It models the acoustic and paralinguistic content of speech so reviewers can examine how something was said, not only what was said. In practice, teams often pair the two: an ASR system for the verbatim transcript and an acoustic classifier for emotion and speaking style. oruk API v1 can return both layers in one response for a prerecorded file.
What acoustic speech models measure
- Emotion & affect
- Joy, anger, doubt, warmth, and the continuous states in between, across speakers and cultures.
- Speaking style
- A separate 16-label multilabel task describing how something was said. The labels characterize delivery, not intent or inner state.
- Sarcastic-sounding delivery
- A mismatch between literal words and vocal delivery. oruk scores how speech sounds, not what the speaker meant.
- Prosody
- Pitch, pace, pauses, and stress, decoded as first-class signal rather than noise.
- Time-local change
- How emotion and speaking-style scores vary across segments of a recorded conversation.
- Frustration
- Tension rising in pitch and pace, surfaced as a calibrated label so reviewers can find those moments.
Why it matters for conversation review
After a conversation ends, transcript-only review can miss frustrated or sarcastic-sounding delivery. Teams can combine transcripts with emotion and speaking-style scores to prioritize completed calls for human review. Voice-agent builders can apply the same analysis to prerecorded evaluation conversations to find responses that need revision. These are post-call review and evaluation workflows, not live routing or de-escalation.
oruk models words and acoustic labels together
oruk API v1 transcribes English audio and returns calibrated multilabel emotion and speaking-style predictions. oruk also publishes measured results through speech-emotion-bench. The production interface is a versioned REST API for prerecorded files.