Pulse — Speech-to-Text by Smallest AI is speech-to-text, stt, asr, real-time transcription, voice ai,. <h1><strong>Pulse: Real-Time Speech-to-Text API for Voice Agents and Enterprise Transcription</strong></h1><p><strong>Pulse</strong>, built by <a target="_blank" rel="noopener noreferrer nofollow" href="https://smallest.ai/speech-to-text">Smallest AI</a>, is a <em>speech-to-text API</em> purpose-built for real-time voice applications: live call transcription, voice agents, and captioning, alongside a high-throughput batch mode for recorded audio. It streams over WebSocket with <strong>sub-100ms time-to-first-token</strong> at low concurrency, and includes speaker diarization, word- and sentence-level timestamps, keyword boosting for domain vocabulary, voice-activity events, and inverse text normalization out of the box.</p><p>A companion model, <a target="_blank" rel="noopener noreferrer nofollow" href="https://docs.smallest.ai/models/model-cards/speech-to-text/pulse-pro">Pulse Pro</a>, trades multilingual support for <strong>English-only batch transcription</strong> tied for <em>#2 on the public Open ASR Leaderboard</em>, built for high-volume call center, meeting, and financial-audio transcription at scale.</p><p>Both models include built-in <strong>PII and PCI redaction</strong> (currently English and Hindi), and the platform holds <a target="_blank" rel="noopener noreferrer nofollow" href="https://security.smallest.ai/?tab=overview">ISO 27001, SOC 2 Type II, HIPAA, and GDPR compliance</a>, with on-premise deployment available on Enterprise plans.</p><h2><strong>Key Features</strong></h2><h3><strong>Real-Time Streaming Transcription</strong></h3><ul><li><p><strong>Sub-100ms time-to-first-token</strong> at 1 concurrency for live voice AI</p></li><li><p>Streams over WebSocket with partial and final transcripts as audio arrives</p></li></ul><h3><strong>Multilingual Speech Recognition</strong></h3><ul><li><p><em>21 languages</em> on streaming, <em>26 languages</em> on batch</p></li><li><p>Regional auto-detect and mid-session code-switching</p></li></ul><h3><strong>Speaker Diarization and Timestamps</strong></h3><ul><li><p>Automatic <strong>speaker diarization</strong> with per-word and per-utterance labels</p></li><li><p>Word- and sentence-level timestamps with per-word confidence scores</p></li></ul><h3><strong>Compliance-Ready Redaction</strong></h3><ul><li><p>Built-in <strong>PII and PCI redaction</strong>, no preprocessing pipeline required</p></li><li><p>Supports regulated industries: healthcare, finance, debt collection</p></li></ul><h3><strong>Custom Vocabulary and Voice Intelligence</strong></h3><ul><li><p><strong>Keyword boosting</strong>: up to 10,000 custom terms per session for brand names, jargon, and product terms</p></li><li><p><em>Emotion and gender detection</em> on batch transcription</p></li></ul><h3><strong>Pulse Pro: Leaderboard Accuracy</strong></h3><ul><li><p>English-only, batch-only speech-to-text model</p></li><li><p>Tied for <strong>#2 on the public Open ASR Leaderboard</strong></p></li></ul><h2><strong>Use Cases</strong></h2><ul><li><p><strong>Voice agents and conversational AI</strong>: real-time transcription for live voice bots</p></li><li><p><strong>Call center and contact center analytics</strong>: transcribe and analyze recorded calls at scale</p></li><li><p><strong>Meeting and financial-audio transcription</strong>: high-accuracy batch processing for enterprise audio</p></li><li><p><strong>Multilingual customer support</strong>: transcribe support calls across 20+ languages</p></li><li><p><strong>Regulated industries</strong>: HIPAA and PCI-compliant transcription with built-in redaction</p></li></ul><h2><strong>Security and Compliance</strong></h2><p>Pulse is backed by Smallest AI's enterprise-grade trust center, covering <a target="_blank" rel="noopener noreferrer nofollow" href="https://security.smallest.ai/?tab=overview">ISO 27001, SOC 2 Type II, HIPAA, and GDPR certifications</a>, with <strong>on-premise deployment</strong> available for teams that need full data residency.</p><h2><strong>FAQ</strong></h2><h3><strong>What languages does Pulse support?</strong></h3><p>Pulse supports 21 languages on real-time streaming and 26 languages on batch transcription, including English, Hindi, Spanish, French, German, Mandarin, Japanese, and Korean, with regional auto-detect for unknown audio.</p><h3><strong>What is the difference between Pulse and Pulse Pro?</strong></h3><p>Pulse handles both streaming and batch transcription across 20+ languages. Pulse Pro is a batch-only, English-only model tuned for maximum accuracy, tied for #2 on the public Open ASR Leaderboard. Use Pulse for live voice agents and multilingual audio, and Pulse Pro for high-volume English transcription where accuracy matters most.</p><h3><strong>Does Pulse support real-time transcription?</strong></h3><p>Yes. Pulse streams transcription results over WebSocket with sub-100ms time-to-first-token at low concurrency, making it suitable for live voice agents, real-time captioning, and conversational AI.</p><h3><strong>Is Pulse HIPAA and GDPR compliant?</strong></h3><p>Yes. The Smallest AI platform holds ISO 27001, SOC 2 Type II, HIPAA, and GDPR certifications, and offers built-in PII and PCI redaction for regulated workflows in healthcare, finance, and debt collection.</p><h3><strong>Can Pulse redact sensitive information automatically?</strong></h3><p>Yes. Pulse includes built-in PII and PCI redaction with no separate preprocessing pipeline required. Redaction currently covers English and Hindi audio.</p><h3><strong>What is keyword boosting used for?</strong></h3><p>Keyword boosting lets you supply up to 10,000 custom terms per session, such as brand names, product names, or domain jargon, so Pulse recognizes them more accurately in the transcript.</p><h3><strong>Does Pulse offer on-premise deployment?</strong></h3><p>Yes. On-premise deployment is available on Smallest AI's Enterprise plan for teams that require full data residency and control.</p><h3><strong>Does Pulse detect emotion or speaker gender?</strong></h3><p>Yes, on batch transcription. Pulse can return emotion detection and gender detection as optional outputs alongside the transcript.</p><hr><p>Learn more or start building at <a target="_blank" rel="noopener noreferrer nofollow" href="https://smallest.ai/speech-to-text">smallest.ai/speech-to-text</a>.</p>. Best for speech-to-text and STT users.
