vsref

Speech-to-text APIs platforms

47 platforms tracked · 983 facts verified · pricing re-checked weekly. Transcription and streaming STT APIs: batch vs realtime cost, accuracy claims with sources, and what diarization actually adds to the bill.

Last verified Jul 24, 2026
Amazon Transcribe logoAmazon Transcribe
AWS-native STT API with deep AWS ecosystem integration and compliance coverage
Alternatives →See pricing
AssemblyAI logoAssemblyAI
Accuracy-led voice AI API for developers and voice agents
Alternatives →See pricing
Azure AI Speech (STT) logoAzure AI Speech (STT)
Enterprise-grade STT inside the Azure cloud ecosystem
Alternatives →From$1,600/mo
Deepgram logoDeepgram
Developer-first realtime STT API for voice agents and transcription at scale
Alternatives →See pricing
ElevenLabs Scribe logoElevenLabs Scribe
Accuracy-led STT API from the leading AI audio company
Alternatives →From$6/mo
Gladia logoGladia
EU-based real-time and batch STT API built on the Solaria models
Alternatives →See pricing
Google Cloud Speech-to-Text logoGoogle Cloud Speech-to-Text
Hyperscaler STT API with Chirp foundation models and enterprise compliance
Alternatives →See pricing
OpenAI gpt-4o-transcribe logoOpenAI gpt-4o-transcribe
Flagship GPT-4o based transcription API from OpenAI
Alternatives →See pricing
Speechmatics logoSpeechmatics
Accuracy-first enterprise STT with flexible deployment (SaaS, container, on-prem)
Alternatives →See pricing
Mistral Voxtral Transcribe logoMistral Voxtral Transcribe
Low-cost EU-based transcription API with Apache-2.0 open-weight models
Alternatives →See pricing
OpenAI Whisper (API) logoOpenAI Whisper (API)
Low-cost pay-as-you-go file transcription from OpenAI; no first-party streaming or diarization
Alternatives →See pricing
Cartesia Ink logoCartesia Ink
Streaming STT for voice agents with native turn detection
Alternatives →From$5/mo
Cohere Transcribe logoCohere Transcribe
Enterprise-grade open ASR for accurate batch transcription
Alternatives →See pricing
xAI Grok Speech-to-Text logoxAI Grok Speech-to-Text
Low-cost hosted STT API on the Grok stack
Alternatives →See pricing
Groq (hosted Whisper) logoGroq (hosted Whisper)
Ultra-fast, low-cost hosted Whisper transcription API (no realtime streaming)
Alternatives →See pricing
IBM watsonx Speech to Text logoIBM watsonx Speech to Text
Enterprise cloud STT API
Alternatives →See pricing
Microsoft MAI-Transcribe logoMicrosoft MAI-Transcribe
Frontier-lab accuracy STT delivered through Azure Speech
Alternatives →See pricing
NVIDIA Parakeet / Riva logoNVIDIA Parakeet / Riva
Open-weights, GPU-accelerated self-hosted STT stack
Alternatives →See pricing
Picovoice (Leopard / Cheetah) logoPicovoice (Leopard / Cheetah)
Private, on-device STT SDK for apps and edge devices
Alternatives →See pricing
Rev AI logoRev AI
Transcription-heritage STT API with low per-hour pricing and open (non-commercial) Reverb models
Alternatives →See pricing
Soniox logoSoniox
Ultra-low-cost multilingual STT + real-time translation API
Alternatives →See pricing
Aqua Voice logoAqua Voice
AI-native system-wide dictation (YC W24) with a developer speech API (Avalon)
Alternatives →From$8/mo
DeepInfra (ASR) logoDeepInfra (ASR)
Low-cost hosted ASR inference
Alternatives →See pricing
Distil-Whisper logoDistil-Whisper
Distilled Whisper: near-Whisper accuracy at a fraction of the size and up to 6x the speed, English-only, MIT-licensed
Alternatives →See pricing
faster-whisper / whisper.cpp logofaster-whisper / whisper.cpp
The de facto local Whisper runtimes: faster-whisper (Python/CTranslate2) and whisper.cpp (C/C++)
Alternatives →See pricing
FunASR / SenseVoice (Alibaba) logoFunASR / SenseVoice (Alibaba)
Open-source industrial ASR toolkit
Alternatives →See pricing
Gradium Speech-to-Text logoGradium Speech-to-Text
Low-latency STT for voice agents
Alternatives →From$13/mo
Handy logoHandy
Private local-first open-source dictation
Alternatives →See pricing
JigsawStack Speech-to-Text logoJigsawStack Speech-to-Text
Whisper-based STT endpoint inside an indie all-in-one small-model AI API platform
Alternatives →From$27/mo
Kyutai STT logoKyutai STT
Streaming-first open STT for self-hosted voice agents
Alternatives →See pricing
Lemonfox.ai logoLemonfox.ai
Cheapest hosted Whisper large-v3 API for batch transcription
Alternatives →From$5/mo
MacWhisper logoMacWhisper
Local-first macOS transcription and dictation app with one-time Pro pricing
Alternatives →See pricing
Monologue logoMonologue
Context-aware Apple dictation by Every
Alternatives →From$14.99/mo
Moonshine logoMoonshine
On-device streaming STT for live voice interfaces, from tiny edge models to Whisper Large v3-beating accuracy
Alternatives →See pricing
Qwen3-ASR logoQwen3-ASR
Open-weights multilingual ASR models
Alternatives →See pricing
Salad Transcription API logoSalad Transcription API
Ultra-low-cost batch transcription on a community GPU cloud
Alternatives →See pricing
Sarvam AI (Saarika / Saaras) logoSarvam AI (Saarika / Saaras)
Indian-language sovereign speech-to-text API
Alternatives →See pricing
sherpa-onnx logosherpa-onnx
On-device ASR/TTS runtime for edge and embedded
Alternatives →See pricing
Smallest.ai Pulse logoSmallest.ai Pulse
Ultra-low-latency multilingual STT for voice agents
Alternatives →See pricing
Superwhisper logoSuperwhisper
AI voice-to-text dictation with context-aware formatting modes
Alternatives →From$8.49/mo
Together AI Transcribe logoTogether AI Transcribe
Low-cost hosted Whisper STT API
Alternatives →See pricing
VoiceInk logoVoiceInk
Open-source, local-first Superwhisper / Wispr Flow alternative with one-time pricing
Alternatives →See pricing
Vosk logoVosk
Lightweight offline STT toolkit for edge and embedded devices
Alternatives →See pricing
WhisperKit (Argmax) logoWhisperKit (Argmax)
On-device Apple Silicon STT (Whisper)
Alternatives →From$1,330/mo
WhisperX logoWhisperX
Whisper + forced alignment + diarization pipeline for accurate word timestamps
Alternatives →See pricing
Willow Voice logoWillow Voice
Cross-platform AI dictation with smart formatting and enterprise-grade privacy
Alternatives →From$15/mo
Wispr Flow logoWispr Flow
System-wide AI voice dictation with auto-editing
Alternatives →From$12/mo

Speech-to-text APIs pricing compared (2026)

Effective monthly bill at 10K min/mo (production app), computed from each vendor's published rates, cheapest first. Every tool page carries the full three-tier table. How we compute costs →

Speech-to-text APIs pricing at 10K min/mo, cheapest first
ToolPublished rateBill at 10K min/moVerified
DeepInfra (ASR)Published rates: batch $0.0004/min · streaming $0.0002/min$4.50Jul 20
Together AI TranscribePublished rates: batch $0.0015/min · streaming $0.0035/min$15Jul 20
SonioxPublished rates: batch $0.0017/min · streaming $0.002/min$16.67Jul 20
xAI Grok Speech-to-TextPublished rates: batch $0.0017/min · streaming $0.0033/min$16.67Jul 20
Groq (hosted Whisper)Published rates: batch $0.0019/min$18.50Jul 20
Qwen3-ASRPublished rates: batch $0.0021/min · streaming $0.0054/min$21Jul 20
SpeechmaticsPublished rates: batch $0.0022/min$21.50Jul 20
Lemonfox.aiPublished rates: batch $0.0028/min$27.78Jul 20
Azure AI Speech (STT)Published rates: batch $0.003/min · streaming $0.0167/min$30Jul 20
Mistral Voxtral TranscribePublished rates: batch $0.003/min · streaming $0.006/min$30Jul 20
Rev AIPublished rates: batch $0.0033/min$33.33Jul 20
Salad Transcription APIPublished rates: batch $0.0033/min$33.33Jul 20
AssemblyAIPublished rates: batch $0.0035/min · streaming $0.0075/min$35Jul 20
ElevenLabs ScribePublished rates: batch $0.0037/min · streaming $0.0065/min$36.67Jul 20
DeepgramPublished rates: batch $0.0043/min · streaming $0.0048/min$43Jul 20
Smallest.ai PulsePublished rates: batch $0.005/min · streaming $0.005/min$50Jul 20
Amazon TranscribePublished rates: batch $0.006/min · streaming $0.01/min$60Jul 20
Microsoft MAI-TranscribePublished rates: batch $0.006/min$60Jul 20
OpenAI gpt-4o-transcribePublished rates: batch $0.006/min$60Jul 20
OpenAI Whisper (API)Published rates: batch $0.006/min$60Jul 20
Aqua VoicePublished rates: batch $0.0065/min · streaming $0.0065/min$65Jul 20
GladiaPublished rates: batch $0.0102/min · streaming $0.0125/min$101.67Jul 20
Google Cloud Speech-to-TextPublished rates: batch $0.016/min · streaming $0.016/min$160Jul 20
IBM watsonx Speech to TextPublished rates: batch $0.02/min · streaming $0.02/min$200Jul 20

What is the lowest-latency STT API?

Smallest.ai Pulse publishes the lowest streaming latency of the 13 Speech-to-text APIs platforms with a figure on record, at 64 ms. Most published latency numbers are the vendor's own benchmark, not an independent one - each row below states which.

Lowest published streaming latency first
ToolStreaming latencySourceVerified
#1Smallest.ai Pulse64 msVendor-claimedJul 20
#2Cartesia Ink88 msVendor-claimedJul 20
#3Moonshine107 msVendor-claimedJul 20
#4AssemblyAI150 msVendor-claimedJul 20
#5ElevenLabs Scribe150 msVendor-claimedJul 20
#6Mistral Voxtral Transcribe200 msVendor-claimedJul 20
#7Willow Voice200 msVendor-claimedJul 20
#8Soniox249 msVendor-claimedJul 20

Popular comparisons

AssemblyAI logoAssemblyAIvsDeepgram logoDeepgram→Deepgram logoDeepgramvsOpenAI Whisper (API) logoOpenAI Whisper (API)→AssemblyAI logoAssemblyAIvsOpenAI Whisper (API) logoOpenAI Whisper (API)→Deepgram logoDeepgramvsElevenLabs Scribe logoElevenLabs Scribe→AssemblyAI logoAssemblyAIvsElevenLabs Scribe logoElevenLabs Scribe→ElevenLabs Scribe logoElevenLabs ScribevsOpenAI Whisper (API) logoOpenAI Whisper (API)→Mistral Voxtral Transcribe logoMistral Voxtral TranscribevsOpenAI Whisper (API) logoOpenAI Whisper (API)→Deepgram logoDeepgramvsMistral Voxtral Transcribe logoMistral Voxtral Transcribe→ElevenLabs Scribe logoElevenLabs ScribevsMistral Voxtral Transcribe logoMistral Voxtral Transcribe→OpenAI gpt-4o-transcribe logoOpenAI gpt-4o-transcribevsOpenAI Whisper (API) logoOpenAI Whisper (API)→Deepgram logoDeepgramvsSpeechmatics logoSpeechmatics→AssemblyAI logoAssemblyAIvsSpeechmatics logoSpeechmatics→Deepgram logoDeepgramvsGladia logoGladia→AssemblyAI logoAssemblyAIvsGladia logoGladia→Deepgram logoDeepgramvsSoniox logoSoniox→Azure AI Speech (STT) logoAzure AI Speech (STT)vsGoogle Cloud Speech-to-Text logoGoogle Cloud Speech-to-Text→Amazon Transcribe logoAmazon TranscribevsGoogle Cloud Speech-to-Text logoGoogle Cloud Speech-to-Text→Amazon Transcribe logoAmazon TranscribevsDeepgram logoDeepgram→Azure AI Speech (STT) logoAzure AI Speech (STT)vsDeepgram logoDeepgram→AssemblyAI logoAssemblyAIvsRev AI logoRev AI→Groq (hosted Whisper) logoGroq (hosted Whisper)vsOpenAI Whisper (API) logoOpenAI Whisper (API)→Cartesia Ink logoCartesia InkvsDeepgram logoDeepgram→NVIDIA Parakeet / Riva logoNVIDIA Parakeet / RivavsOpenAI Whisper (API) logoOpenAI Whisper (API)→AssemblyAI logoAssemblyAIvsSoniox logoSoniox→Deepgram logoDeepgramvsGoogle Cloud Speech-to-Text logoGoogle Cloud Speech-to-Text→Gladia logoGladiavsOpenAI Whisper (API) logoOpenAI Whisper (API)→Deepgram logoDeepgramvsRev AI logoRev AI→Deepgram logoDeepgramvsNVIDIA Parakeet / Riva logoNVIDIA Parakeet / Riva→Moonshine logoMoonshinevsOpenAI Whisper (API) logoOpenAI Whisper (API)→Qwen3-ASR logoQwen3-ASRvsOpenAI Whisper (API) logoOpenAI Whisper (API)→Cohere Transcribe logoCohere TranscribevsDeepgram logoDeepgram→Deepgram logoDeepgramvsxAI Grok Speech-to-Text logoxAI Grok Speech-to-Text→Deepgram logoDeepgramvsSmallest.ai Pulse logoSmallest.ai Pulse→Deepgram logoDeepgramvsIBM watsonx Speech to Text logoIBM watsonx Speech to Text→

Best for…

Call CentersAzure AI Speech (STT)DevelopersAzure AI Speech (STT)DictationMoonshineMedicalAssemblyAIMeetingsAssemblyAISelf-HostedMistral Voxtral TranscribeVoice AgentsCartesia Ink

Want it done for you?

Most teams pricing an STT API for meetings do not actually want to build the bot, the diarized transcript, and the summary themselves. AI meeting notetakers ship that whole pipeline: Otter.ai, Fireflies.ai, Fathom, or the full notetaker comparison.

Voice stack cost calculator →See the real combined cost per minute of a platform, an STT API, and a TTS voice from Speech-to-text APIs and its sibling categories.

More verified comparisons

AI app builders →AI app builders compared: what a credit actually buys, which free tiers survive a real build, and the self-hosted options nobody tables.AI avatar video →AI avatar video platforms: what a finished minute actually costs once you decode the credits, and who owns the output.AI coding agents →AI coding agents compared: what $20 a month actually buys once the usage limits kick in, which models each tool can run, and the open-source agents nobody tables.AI meeting notetakers →AI meeting notetakers compared: bot vs bot-free, what free tiers actually cap, and the real per-seat cost for a team.Social media scheduling →Social media scheduling platforms compared: what per-channel pricing really costs at your channel count, which free plans survive real use, and the self-hosted option nobody tables.Text-to-speech APIs →APIs and platforms that turn text into speech: real per-minute cost, latency, cloning rights, and a verdict per use case.Voice AI →Voice AI platforms that make and answer phone calls: verified pricing, real production cost per minute, and a verdict per use case.