
1 / 3
【Overview】 Pulse is a speech-to-text API built for developers who need to turn live speech, calls, audio and video into structured text. It streams audio over a persistent WebSocket connection and returns partial and final transcript events as the model processes incoming audio, so applications can react to speech while it is still being spoken. The API covers both live and recorded audio. For files, Pulse accepts audio and returns a complete transcript in a single synchronous HTTP response. On top of the raw text, Pulse adds speech intelligence: speaker diarization, emotion detection, language detection, PII and PCI redaction, and per-word timestamps. Pulse is aimed at developers building real-time speech pipelines, contact centers, voice AI applications, and production workloads at scale. It transcribes speech in 38+ languages across accents, dialects, and recording conditions. 【Key features】 • Real-time transcription: Streams audio over a persistent WebSocket connection and returns partial and final transcript events as the model processes incoming audio. • Pre-recorded transcription: Accepts audio files and returns a complete transcript in a single synchronous HTTP response. • Speaker diarization: Identifies transitions between speakers and accurately labels each contribution. • Emotion detection: Detects user emotions to make conversations more empathetic. • Language detection: Automatically identifies the language in incoming audio before downstream processing. • PII / PCI redaction: Handles sensitive information in transcripts before it moves into downstream workflows. • Multi-language transcription: Transcribes speech in 38+ languages across accents, dialects, and recording conditions. • Word timestamps: Provides per-word timing useful for caption generation, subtitle tracks, and confidence-based review workflows. 【Use cases】 1. Build voice agents 2. Live captions 3. Meeting notes 4. Subtitles 5. Call analytics 【Pricing】 Pay as you go: $0.003/minute — Pulse (Pre-Recorded) ~$0.003/minute, Pulse (Realtime) ~$0.004/minute, Pulse Pro (Pre-Recorded) ~$0.0035/minute, 100 concurrent streams Enterprise Plan: Custom — Pulse on Prem, Custom concurrency for STT, Enterprise Grade, 99.99% SLA, Priority Support 【Pros and cons】 Pros: - The usage-based pricing lets teams start small and scale spend with actual audio volume rather than committing to a fixed plan. - Partial and final transcript events over a WebSocket make it a natural fit for products that must respond to speech in the moment, such as voice agents. - Built-in speaker labels, language detection, and PII/PCI redaction save teams from assembling separate post-processing steps. 【FAQ】 Q: How is Pulse priced? A: Pulse uses pay-as-you-go pricing at $0.003/minute, with pre-recorded and realtime variants priced separately and a custom Enterprise Plan available. Q: Who is Pulse built for? A: Developers building real-time speech pipelines, contact centers, voice AI applications, and production workloads at scale. Q: Can Pulse transcribe languages other than English? A: Yes, Pulse transcribes speech in 38+ languages across accents, dialects, and recording conditions, and can detect the language automatically.
Pay as you go · $0.003/minute (Pulse (Pre-Recorded) ~$0.003/minute, Pulse (Realtime) ~$0.004/minute, Pulse Pro (Pre-Recorded) ~$0.0035/minute) Enterprise Plan · Custom (Pulse on Prem, Custom concurrency for STT, Enterprise Grade, 99.99% SLA)
WPS AI is an AI assistant built into WPS Office that helps users write documents, generate presentation outlines and…
Baidu Comate is an AI coding assistant built by Baidu on its ERNIE model. It goes beyond simple autocomplete: it…
UniConverter is an AI media conversion and enhancement tool for video, audio, and images. It combines everyday…
6 similar tools to consider instead of Pulse