Skip to main content

Speechly features

Feature availability varies depending on if you are using pre-recorded audio or live streaming audio, your application language and your project payment plan.

CategoryFeaturePre-recordedLive streamingAvailability
ModelsConformer RNN‑T models33
Whisper models1
Model adaptation / trainingRNN‑T models only
Data annotation serviceEnterprise plans only
TranscriptionPre-recorded audio
Live streaming audio
Language support99English
Language detection
Model selection
Word level timestampsRNN‑T models only
Punctuation
Number & date formatting
Silence segmentationRNN‑T models only
Interim resultsRNN‑T models only
Voice activity detection
Evaluate ASR accuracy
Speech understandingIntent detectionRNN‑T models only
Entity detectionRNN‑T models only
LookupsRNN‑T models only
Text labelingEnglish language only
Language translationWhisper models only
Evaluate NLU accuracy
Audio analysisLanguage detection
Audio event labeling
Tone of voice
Lyrics transcriptionEnglish language only
Supported audio formatsWAV
FLAC
OGG
MP3
AAC??
Up and down sampling
Deployment optionsOn-deviceEnterprise plans only
On-premiseEnterprise plans only
Cloud
IntegrationBrowser client
React client
Android client
iOS client
Unity client
Speechly decoderiOS, Android and C
gRCP API
REST API
Developer toolsDashboard
CLI

✓ available   * planned   ? under consideration