Lightning-fast speech to text

We power leading providers of customer voice analytics


Rapid transcription of all live and recorded calls with lowest total cost of ownership


Best in class ASR technology tuned to specific business needs


Protects the privacy of customer audio and transcripts, so you control your data


Full punctuation, speaker separation, gender, acoustic emotion and redaction


Integrates seamlessly with your stack, call provider, and telephony system


AI / Deep Learning

Discover actionable insights

Voci proprietary machine learning and deep neural networks ingest and “learn” voice data. Our innovations determine customer satisfaction through sentiment analysis and speech characteristics, and recognize demographic-based trends through gender identification.


Learn more about our AI and deep learning capabilities.

Speech to Text

APIs and Integration

Combine transcripts and analytics

Our V-Blaze speech engine features industry-standard open APIs that are compatible with over 350 telephony audio formats. Its open-format JSON and Text transcripts integrate with our
V-Spark speech browser and third-party analytics or business intelligence platforms.

Find out more

Speech to Text

Speech to Text

Put it in writing

Voci speech-to-text technology uses GPU acceleration to process 100% of live and recorded calls into highly accurate transcripts. This enables you to improve agent quality monitoring, extract competitive intelligence, and enhance customer experience.

Our Technology



Hear the true voice of every customer

  • NVIDIA® GPUs enable lighting-fast transcription of live and recorded calls
  • Highly accurate transcripts integrate with your analytics platform to reveal actionable insights and the true voice of every customer

V-Blaze transcripts provide:

  • Full punctuation
  • Gender, emotion, and sentiment insights
  • Event-level metadata including timestamps
  • Numeric redaction of audio, text, or both
  • Automatic speaker separation (diarization)


Visualize customer calls with a powerful speech browser

  • Reveals emotion and sentiment
  • Classifies your data for immediate business action
  • Integrates easily with analytics or business intelligence platforms

V-Spark speech browsing provides:

  • An intuitive web-based interface
  • Advanced search, filter, playback, and tag capabilities
  • Drill-down to specific call types
  • Customizable applications for reporting and discovery
  • Importing of external metadata (e.g., agent ID, supervisor) for filter and sort capabilities


Verify speaker identity at any point during a call

  • Enables partners to develop and integrate advanced voice biometrics technology in their current processes
  • Integrates easily with telephony audio and third-party database solutions
  • Improves fraud prevention, customer experience, and public safety

V-Match verification provides:

  • Passive voice matching, which enables flexible and complete security solutions
  • Batch and incremental processing, in both offline and real-time settings
  • An open API architecture to develop customized voiceprint creation and matching applications

Stay updated with Voci's speech insights

Please type your first name.
Please type your last name.
Invalid email address.
Invalid Input
I have read and agree with Voci’s Privacy Notice

Become a partner

Voci greatly values our partners’ expertise. Learn how you can partner with us as a platform provider, value-added reseller, or service provider and increase your business opportunities.