Products
A complete speech and language technology portfolio.
Three product families cover recognition, intelligence and interaction. Every product supports private deployment options; not every product supports every option. Individual product pages describe what is available.
Product family
Speech Recognition & Transcription
Turn spoken language into accurate, usable text with models adapted to specialized environments and terminology.
View Speech Recognition & Transcription →Cobalt Transcribe
Automatic speech recognition (ASR) for batch and streaming transcription.
Cobalt Transcribe Tuner
Adapt acoustic and language models to your domain, vocabulary and audio conditions.
Cobalt Voice Channel
Speech recognition tuned for telephony and contact-center audio.
Cobalt Detect
Detect specific keywords, phrases or events within audio streams.
Product family
Voice Intelligence
Protect sensitive information, authenticate speakers, and analyze characteristics contained in voice.
View Voice Intelligence →Cobalt Privacy Screen
Detect and redact PII and other sensitive content from audio and transcripts.
Cobalt VoiceBio
Voice biometrics for speaker verification and identification.
Cobalt Speech Intelligence
Extract structure and meaning from spoken conversations.
Product family
Voice User Interfaces
Build natural spoken interactions for applications, devices, portals, equipment and support systems.
View Voice User Interfaces →Cobalt Interact
Conversational voice interface framework combining ASR, NLU and TTS.
Cobalt VoiceGen
Text-to-speech (TTS) with configurable voices.
Cobalt Understand
Natural language understanding (NLU) for spoken and typed input.
Cobalt Interact Model Updater
Update deployed interaction models without full re-releases.
Product family
Cobalt Labs
Custom research and development for speech and language problems that off-the-shelf models do not address.
View Cobalt Labs →Custom Speech R&D
Applied research targeting your specific data, environment and constraints.
Choosing
Which family fits your problem?
| If you need to… | Start with |
|---|---|
| Convert audio to text at scale | Speech Recognition & Transcription |
| Redact sensitive data from audio or transcripts | Voice Intelligence — Privacy Screen |
| Authenticate speakers by voice | Voice Intelligence — VoiceBio |
| Build a spoken interface for a device or application | Voice User Interfaces — Interact |
| Generate synthesized speech | Voice User Interfaces — VoiceGen |
| Solve a problem that off-the-shelf models miss | Cobalt Labs |