Compare realtime APIs

Pick up to four APIs from any category and see them side by side.

Azure AI Speech real-time speech to text
Microsoft
ElevenLabs Scribe v2 Realtime
ElevenLabs
Voxist ASR (via OVHcloud / Scaleway marketplaces)
Voxist
CategorySpeech-to-textSpeech-to-textSpeech-to-text
StatusGAGAGA
Est. per minute$0.0067 - 0.025$0.0065 - 0.0098$0.0063 - 0.016
How that was worked out$0.40/hr commitment-tier overage up to $1.20/hr custom + $0.30/hr add-on. Pay-as-you-go standard is $0.0167/min.$0.39/hr base to $0.59/hr with keyterms + editing.Licence EUR 0.35-0.88/hr (~$0.006-0.016/min at ~1.1 USD/EUR) before GPU compute. Low confidence.
Pricing modelper-hourper-hourper-hour
Free tierF0: 5 audio hours per month shared between standard and custom real-time; 1 concurrent request.Free plan includes about 2.5 realtime hours.None stated.
Connects byWebSocket (via Speech SDK), REST for short audioWebSocketWebSocket, REST
Audio inSpeech SDK handles microphone, files and push/pull streams; default 16 kHz 16-bit mono PCM; compressed formats via GStreamer.audio_format: pcm_8000, pcm_16000 (default), pcm_22050, pcm_24000, pcm_44100, pcm_48000, ulaw_8000; audio sent base64 in JSON input_audio_chunk messages.WAV, mono only; examples use 16 kHz.
Audio outn/an/an/a
Languages100+ locales (see language support page; count not re-verified).90+ languages (vendor); language_code plus secondary_languages hints.Not listed on the billing page (examples show en and French output).
Latency (vendor claim)No vendor figure captured.Vendor claim: ~150 ms (footnoted, conditions not stated).Not captured.
Key limits
  • Concurrent real-time requests: F0 1 (fixed), S0 100 default (adjustable); limit is shared with speech translation.
  • Diarization identifies up to 35 speakers (errors beyond that).
  • Realtime concurrency limit by plan is in a separate chart on the models page (not captured).
  • keepalive_interval_ms configurable 500-10000; errors include session_time_limit_exceeded, commit_throttled, queue_overflow, insufficient_audio_activity.
  • WAV mono only.
  • Streaming has no diarization.
High-severity warnings
  • Pricing page shows no numbers
  • None
  • None
ComplianceAzure compliance scope (HIPAA BAA, SOC, ISO) applies to Azure AI Speech per Microsoft; connected/disconnected containers available. Not re-verified.Data-residency hosts for EU, India, Singapore; enable_logging=false for zero retention. Certifications not re-verified.Runs in your own cloud project; certifications not checked.
Self-hostableYesNoYes
Last checked2026-10-102026-10-102026-10-10
Key numbers and features
$/hour$1$0.39-
Bills silence---
Free tierYesYes-
Free credit $---
Live speakersYes-No
Turn detect-Yes-
KeytermsYesYes-
PII redact---
PartialsYesYes-
Mixed langsYes--
8 kHz phone-YesNo
Latency ms-150-
Languages-90-
Max session min---
Concurrency1009-
WebRTCNoNoNo
WebSocketYesYesYes
gRPCNoNoNo
HIPAAYes--
SOC 2Yes--
EU dataYesYesYes
Self-hostYesNoYes
Open weightsNoNoNo
High warnings100