Skip to main content

Automatic Speech Recognition (ASR)

We support the following third-party service providers for ASR services:
ASROn-Prem / CloudLanguagesRegionsWord Error Rate (WER)Comments
GoogleCloudSupported Languages- Locations v2 - Docs
- Regions
4-9%- Good for short utterances (for example, “yes”, “no”).
- Works well for numeric and alphanumeric inputs (IDs, SSN).
- Supports class tokens for output formatting.
- Supports hints and hint-boosts.
- Extensive language support.
DeepgramCloud & On-PremSupported LanguagesSupports all regions globally3.44%- Supports hints.
- Provides custom models via Deepgram team.
- Supports smart formatting (numbers, dates).
AzureCloud & On-PremSupported LanguagesRegions5-10%- Preferred ASR provider; default for new accounts.
- Low WER with flexible customization.
- Supports hints.
- Extensive language support.
- Supports custom model creation via Azure portal.
Nvidia Riva (Nvidia)On-PremASR Overview-- 67%
Amivoice ASR (Advanced Media Inc)CloudSupported LanguagesPrimarily Japan-based processing and storageN/A
Amazon TranscribeCloudSupported LanguagesRegions- 60%
gnani.aiCloud & On-PremSupported LanguagesDeployable in customer-specified regions (private cloud or on-premises)2%

Text to Speech (TTS)

We support the following third-party service providers for TTS services:
TTSOn-Prem / CloudLanguagesRegionsComments
GoogleCloudSupported VoicesOperates within Google Cloud’s global infrastructure
AzureCloud & On-PremSupported LanguagesRegions- Extensive language support.
- Large number of voices.
- Supports custom voice creation through the portal.
- Supports SSML (limited to Azure-supported tags).
OpenAI TTSCloudSupported Languages-- Human-like voices.
- Limited number of voices.
Eleven LabsCloudDocs-- Human-like voices.
- Supports speed, temperature, and stability controls.
- Supports voice cloning with 30-60 second samples.
AWSCloudSupported LanguagesRegions
gnani.aiCloud & On-PremAPI Service-
DeepgramCloud & On-PremSupported Languages-- Limited number of languages.
- Human-like voices.
Nvidia Riva TTSOn-Prem---

Voice Biometrics

We support the following third-party service providers for voice biometrics:
Voice Biometric VendorVoice Biometric EngineOn-Prem / CloudComments
ID R&DID Voice--