# Smallest.ai Pulse STT > Smallest AI's Pulse speech-to-text API transcribes pre-recorded files over HTTP and live audio over WebSocket, with three models (Pulse, Pulse 2.0 and Pulse Pro), diarisation, word timestamps, redaction and turn detection. - Canonical: https://www.anchorterminal.com/tools/smallest-pulse-stt - Pricing: https://www.anchorterminal.com/tools/smallest-pulse-stt#pricing - Reviews: https://www.anchorterminal.com/tools/smallest-pulse-stt#reviews - Markdown: https://www.anchorterminal.com/tools/smallest-pulse-stt.md (~8,400 tokens) - Slim: https://www.anchorterminal.com/tools/smallest-pulse-stt.min.md (~1,780 tokens, same facts, less prose, for token-sensitive contexts) - JSON: https://www.anchorterminal.com/tools/smallest-pulse-stt.json (this page as data, same URL with Accept: application/json) - Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt) - API: https://www.anchorterminal.com/api/v1/index.json - MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key) - DNS: dig +short TXT smallest-pulse-stt.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.. and .vs.. under the zone, and the same under grade.letme.dev) - Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data) - Updated: 2026-11-07 ## Overview **Grade B · 65/100 · rank #441 of 1486 · #13 in Speech-to-text · not agent-ready · confidence medium** More from Smallest AI, listed separately because each is its own product: [Smallest.ai Lightning TTS](https://www.anchorterminal.com/tools/smallest-lightning-tts.md) (Text-to-speech), [Smallest.ai Instant Voice Clone](https://www.anchorterminal.com/tools/smallest-voice-clone.md) (Voice cloning & custom voices). ## Assessment Pulse covers files and live audio on one endpoint, with OpenAPI and AsyncAPI files, a dated changelog and public per-minute prices from $0.003. The shared API's health checks failed in every region on five days in September, keys carry no scopes and can travel in a URL, and zero retention is Enterprise only. ## Facts | Field | Value | | --- | --- | | Vendor | Smallest AI (https://smallest.ai) | | Kind | Model API | | Category | Speech-to-text (https://www.anchorterminal.com/categories/speech-to-text) | | Transport | HTTP, websocket | | Auth | API key · Bearer `sk_` API key created in the dashboard after a browser signup, shared by every model and the Voice Agents API. Keys are shown once, revoked by deleting them and carry no scopes. Servers can mint `wat_` access tokens for browsers (30 to 900 seconds, inference routes only), which the STT socket takes as `?api_key=`. | | Pricing | Pay per use (Pay per use) · Pay as you go from $0.003 a minute for Pulse batch, $0.004 for Pulse streaming and $0.005 for Pulse 2.0. Pulse Pro is listed at both $0.0035 and $0.004 a minute. $10 in trial credits. Enterprise pricing, SLA and zero data retention by quote. | | x402 | No · No x402, MPP or L402 in the docs, OpenAPI file or pricing page (checked 2026-11-06). Credits are bought with a card on file. | | Licence | unknown | | Tools exposed | 0 | | Packages | pypi: `smallestai`; npm: `smallestai`; npm: `@developer-smallestai/smallest-mcp-server` | | Source | https://github.com/smallest-inc/smallest-python-sdk | | Docs | https://docs.smallest.ai/models/speech-to-text/overview | | llms.txt | https://docs.smallest.ai/llms.txt | | Last release | 2026-11-02 | | Surface listed | The Pulse speech-to-text routes of the Models API at api.smallest.ai/waves/v1, read from docs.smallest.ai (Markdown pages, llms.txt, OpenAPI and AsyncAPI files, changelog), smallest.ai pricing, terms and privacy pages and status.smallest.ai on 6 November 2026. No account was made and no request was sent to the API | | Models | `pulse` (multilingual, streaming and batch), `pulse-2` (English streaming with end-of-turn, emotion and gender) and `pulse-pro` (English batch only, which the vendor says is tied second on the Open ASR Leaderboard at 5.42% WER) | | Languages | Streaming takes 21 single-language codes plus `north_indic`, `multi-asian` and `multi-south-indic`. Batch takes 12 codes plus `multi-eu`, `multi-asian` and `multi-indic`. Eleven languages are labelled Beta, and East Asian streaming runs from the US region | | Options | Word timestamps, utterances, diarisation with integer speaker labels, PII and PCI redaction (Pulse only), keyword boosting up to 100 terms, emotion and gender detection, inverse text normalisation, VAD events and webhooks for batch results | | Regions | Mumbai and Oregon, routed by proximity, with region-pinned hosts api.india.smallest.ai and api.us.smallest.ai | | MCP server | `@developer-smallestai/smallest-mcp-server` 0.11.2 (MIT, npm) runs locally with npx and mostly manages voice agents, with a `transcribe_audio` tool among about 80 | | Other products | Lightning text to speech (listed separately), Hydra speech to speech, the Electron LLM, voice cloning and the Atoms voice agent platform share the same key and quota | | SDKs | Python `smallestai` 5.14.0 (7 October 2026). The npm `smallestai` Node SDK is at 1.0.1, last published in May 2025 | | Capabilities | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | | Tags | hosted, usage, closed-source, python, typescript, openapi, llms-txt, streaming, batch, webhooks, mcp, enterprise, self-hosted | | JSON | https://www.anchorterminal.com/api/v1/tools/smallest-pulse-stt.json | ## Score breakdown (methodology v0.7, October 2026 research run) Assessed 2026-11-06 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: medium. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. "This run" is each category's share of the 100 points. | Category | Weight | This run | Score (0–100) | Points | | --- | --- | --- | --- | --- | | Reliability | 16% | 20 | 63 | 12.6 | | Performance | 10% | pending | pending | n/a | | Schema & documentation | 13% | 16.2 | 92 | 14.9 | | Agent ergonomics | 13% | 16.2 | 66 | 10.7 | | Security & auth | 14% | 17.5 | 52 | 9.1 | | Payments & pricing | 10% | 12.5 | 40 | 5.0 | | Task success | 10% | pending | pending | n/a | | Maintenance & community | 7% | 8.8 | 73 | 6.4 | | Transparency & trust (editorial 55, provenance 87) | 7% | 8.8 | 71 | 6.2 | | Negative events | up to −15 | up to −15 | none recorded | 0 | | **Total** | | | | **65 → B** | ### Why each score - Reliability 63 (checked 2026-11-06): BetterStack status page at status.smallest.ai with components for the Global, India and USA APIs (one health check covers the Agents and Models APIs), docs, the app and Electron, and per-component history (20). In the 90 days from 8 August the health checks on the shared API failed in all three regions on 9, 10, 13, 14 and 15 September (about 8, 21, 33, 17 and 64 minutes of flapping), in the Global and India regions on 11 and 17 September (12 minutes each), and in the USA region alone on 1, 3, 5, 12, 21 and 30 September (6 to 62 minutes) and 6 October (12 minutes). Outages in every region at once are read as wide errors, so several majors (0). Standard plan limits published per model, 100 concurrent streaming sessions and 25 batch requests a minute, and the pricing page also says 100 concurrent STT streams (15). 429 is documented with exponential backoff from 500 ms to 30 s and Retry-After when present, and batch calls are synchronous with optional webhooks (15). An infrastructure SLA of 99.99 per cent is named for Enterprise only, with no terms published, and the terms disclaim any availability commitment (5 of 10). Pulse, Pulse 2.0 and Pulse Pro carry no preview label, but eleven language codes are marked Beta (8 of 10). 20 + 0 + 15 + 15 + 5 + 8 = 63. - Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes. - Schema & documentation 92 (checked 2026-11-06): OpenAPI 3.1 for the HTTP routes and AsyncAPI 2.6 for the STT socket, with query parameters and event schemas (25). llms.txt and a Markdown copy of every page (10). Model cards and a compare page say which model suits which audio, and fields say where they're ignored (Pulse Pro ignores redaction), but several changelog entries correct response fields the reference had wrong, such as the diarisation shape and a `redacted_entities` array that doesn't exist (15 of 20). Typed query parameters with enums, defaults and required marks, though booleans are string enums that must be lowercase (13 of 15). Examples in curl, Python and JavaScript, and the reference lists 400, 401, 403, 413, 422 and 429 bodies with known `code` values (14 of 15). Versioned `/v1` paths, a deprecations page and a dated Pulse changelog (15). Total 92. - Agent ergonomics 66 (checked 2026-11-06): API reading of the checklist. Word timestamps, utterances, diarisation, emotion and gender are each opt-in, so the default response is the transcript with metadata (18 of 25). There are no list endpoints for transcriptions to page. Output controls are the opt-in fields and finalisation settings on the socket (10 of 20). Errors carry a status and a machine-readable `code` or `error_code` (`LANGUAGE_NOT_SUPPORTED_BY_MODEL`, `AudioTooLong` and others) with fixes, but the body comes in two shapes, `message` or `error` (15 of 20). Batch calls are synchronous with no idempotency key, and whether a retried file is billed again isn't stated (10 of 20). A file call needs only the model and audio. Official Python and Node SDKs exist, though the Node SDK was last published in May 2025 (13 of 15). Total 66. - Security & auth 52 (checked 2026-11-06): Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Plain `sk_` keys per organisation, shown once, revoked at once and deletable, with one key per environment advised. The docs say a key grants full account access, and no scope is offered. Short-lived access tokens (`wat_`, 30 to 900 seconds) cover only inference routes and can't mint tokens (22 of 30). The error reference documents the key as a `?token=` query parameter for browser WebSockets, and the token guide passes access tokens as `?api_key=` on the STT socket, so 10 off (12 of 30). The terms (21 October 2026) say Smallest won't use user content or input to train or improve its services. The privacy policy (18 September 2026) lists improving the services with machine learning as a use of usage data and doesn't address uploaded audio or text either way (15 of 20). Zero data retention and the `x-expire-content` header are Enterprise only, or a $1,000 a month HIPAA add-on on pay as you go. Other plans keep content readable in logs and the dashboard, with no retention period published (5 of 15). Usage is tracked per key, request rows sit in a dashboard log viewer, and every response carries `x-request-id`. No audit log of key actions was found (10 of 15). security.txt is valid to 31 December 2026 and the site footer claims SOC 2, GDPR and HIPAA compliance, while the pricing page lists SOC 2 compliance only on Enterprise. No bug bounty, disclosure policy or public advisories found (10 of 20). Total 52. - Payments & pricing 40 (checked 2026-11-06): No x402, MPP or L402 (0). Per-minute prices for each Pulse model published without a login, with two figures for Pulse Pro (20). Pay as you go starts with $10 of trial credits, which the terms also describe, and no page says a card is needed to claim them, read as inworld-tts's free plan was (20). A person signs up in a browser at app.smallest.ai and creates the key in the dashboard, and top-ups go to a card on file (0). Total 40. - Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored. - Maintenance & community 73 (checked 2026-11-06): Read as a model. The newest Pulse changelog entry is dated 2 November 2026 (updated Indic benchmarks), and Pulse 2.0 launched within the period (30). Dated Pulse entries on 30 August, 18, 29 and 30 September, 1, 12, 13 and 21 October and 2 November (20). Public changelog, email support and a Discord server, with no evidence read of how they answer (10 of 15). The Python SDK `smallestai` 5.14.0 dates from 7 October with 12 releases since 9 August, but the npm Node SDK has not been published since May 2025 (8 of 15). The Python SDK repository runs CI and release workflows, but has no licence file (5 of 10). Total 73. - Transparency & trust 71 (checked 2026-11-06): Closed service with clear terms naming Smallest Inc. under California law (15). The ZDR page is specific about what each mode keeps, but the terms' no-training clause and the privacy policy's machine-learning use of usage data sit side by side, the privacy policy gives no retention periods, and the policy doesn't mention the DPA a footer link points to (15 of 30). A deprecations page gives statuses, dates and migration targets, deprecated routes return a `Deprecation: true` header, and the 30 September changelog removed 14 European codes from the batch enum with a migration to `multi-eu`. The terms still let Smallest change or discontinue services at any time without notice (15 of 20). Two regions named (Mumbai and Oregon) with region-pinned hosts, and only Google Analytics and Fullstory named among third parties, with no sub-processor list found (10 of 20). Total 55. Fix list, everything this grade says the listing lacks, the biggest gain first (16 items). What it covers is at https://www.anchorterminal.com/fixes/smallest-pulse-stt.md (JSON https://www.anchorterminal.com/fixes/smallest-pulse-stt.json). The whole list for a coding agent comes from `POST https://www.anchorterminal.com/api/v1/fix-list` with `{"listing": "smallest-pulse-stt", "plan": "read"}`, free while the beta runs ### What we couldn't check - PyPI's JSON API (pypi.org/pypi/smallestai/json), which pypi.org/robots.txt disallows, was fetched once by mistake on 6 November. Release dates were re-read from the allowed RSS feed, and nothing taken only from the JSON is used. - unchecked: whether signup asks for a card before the $10 trial credits - unchecked: open issues and maintainer replies on the SDK repositories, because the GitHub API refused our requests for its rate limit - The pricing page gives two prices for Pulse Pro, $0.0035 and $0.004 a minute - unchecked: whether a retried or failed transcription is billed ### Sources - status page and incident history (JSON): (seen 2026-11-06) - concurrency and limits: (seen 2026-11-06) - model pricing: (seen 2026-11-06) - plans and free credits: (seen 2026-11-06) - terms of service: (seen 2026-11-06) - privacy policy: (seen 2026-11-06) - zero data retention: (seen 2026-11-06) - API keys and access tokens: (seen 2026-11-06) - Models API OpenAPI 3.1: (seen 2026-11-06) - Models API AsyncAPI 2.6: (seen 2026-11-06) - model error reference: (seen 2026-11-06) - model deprecation notices: (seen 2026-11-06) - Python SDK releases on PyPI: (seen 2026-11-06) - security.txt: (seen 2026-11-06) - Pulse STT overview: (seen 2026-11-06) - Pulse model card: (seen 2026-11-06) - Pulse changelog: (seen 2026-11-06) - transcribe API reference: (seen 2026-11-06) ## Score history Every move of a score since the listing was graded, newest first, with its cause. A score can't move without an entry here. JSON: https://www.anchorterminal.com/api/v1/tools/smallest-pulse-stt/history.json - 2026-11-06: graded 65 B (methodology 0.7) ## Who's behind it (provenance 87/100, checked 2026-11-06) | Check | Finding | Points | | --- | --- | --- | | Legal entity named | Smallest, Inc. | 20/20 | | Domain age | smallest.ai, registered 2023-09-02 (3 years) | 7/15 | | Endpoint on the vendor's domain | smallest.ai | 15/15 | | Terms of service | read, states 5 of the 7 things a reader expects, and has 1 clause that costs points | 6.3/10 | | Privacy policy | read, states 5 of the 8 things a reader expects | 7.8/10 | | Status page | status.smallest.ai | 10/10 | | Changelog | published | 10/10 | | security.txt | valid | 10/10 | | People named | 2 named with their role, none with a public profile | 7/10 | | Support route | a support address, help pages, a community | 5/5 | The terms of service (effective 21 October 2026) name Smallest Inc., 311 California Street, Suite 320, San Francisco, under California law. The privacy policy (effective 18 September 2026) names Smallest, Inc. with a registered office in Lewes, Delaware. The terms forbid accessing or searching the services with spiders, robots, crawlers or similar data mining tools except through means Smallest makes available. Only public pages were read, with no account. smallest.ai/robots.txt allows all crawlers including ClaudeBot, with Content-Signal ai-train=yes, search=yes, ai-input=yes. docs.smallest.ai/robots.txt allows all paths except /api/fern-docs/. /.well-known/security.txt names security@smallest.ai and expires 31 December 2026. /security.txt returns 404. RDAP gives smallest.ai a registration date of 2 September 2023. Every docs page served as Markdown opens with a block addressed to AI assistants that says which models to prefer. It was read as data, not followed. firstReleased is left empty. No dated launch of the Models API was found on a page our reader may fetch. ### Terms and privacy, as read A reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice. **Terms of service** (https://smallest.ai/terms-of-service), read 2026-11-06, gives no date, states 5 of the 7 things a reader expects. - To know. Says the terms or the service can change without notice (costs points). "Because our Services are evolving over time we may change or discontinue all or any part of the Services, at any time and without notice, at our sole discretion." - To know. Says access can be ended without notice or for any reason. "We may suspend or terminate your access to and use of the Services, including suspending access to or terminating your Account, at our sole discretion, at any time and without notice to you." - To know. Requires arbitration or waives class actions. "(a) Mandatory Arbitration of Disputes." - Not found in the text. Gives the date it was last updated. - Names the governing law or courts. The law of Federal Arbitration Act. - States a limit on its liability. Capped at $100.00. - Says how changes to the terms are announced. Gives 30 days of notice before a change. - Not found in the text. Refers to a service level or uptime commitment. **Privacy policy** (https://smallest.ai/privacy-policy), read 2026-11-06, gives no date, states 5 of the 8 things a reader expects. - Not found in the text. Gives the date it was last updated. - Says how long data is kept. For as long as needed, with no period named. - Not found in the text. Says whether personal data is sold or shared for advertising. - Not found in the text. Gives a privacy contact. ## Humans behind the project (researched 2026-11-06 from public pages, not part of the grade) ### Key people - Sudarshan Kamath, Co-founder and CEO - Akshat Mandloi, Co-founder and CTO - Note: Smallest AI's press kit, read 6 November 2026, names Sudarshan Kamath as co-founder and CEO and Akshat Mandloi as co-founder and CTO, and says the two worked together on Bosch's autonomous vehicles team. The press kit's portrait of the CTO carries another person's name in its alt text. The privacy policy names Sudarshan Kamath as grievance officer. No personal accounts are linked from Smallest AI's pages, so none are listed. Investors come from Smallest AI's own Series A post of 5 October 2026 ($13M, led by Seligman Ventures, total funding above $21M, with unnamed angels) and from SiliconANGLE's report of 9 October 2025 on the $8M seed led by Sierra Ventures, with 3one4 Capital and Better Capital. Inc42's report of the seed, cited before, was dropped on 6 November 2026 because inc42.com's robots.txt shuts out Anthropic's crawlers, so Upsparks Capital, Schema Ventures and DeVC now carry only the Series A, where Smallest's post names them. A web search on 6 November 2026 found no valuation stated for either round. - Smallest AI (Smallest, Inc.): LinkedIn https://www.linkedin.com/company/smallest/ X https://x.com/smallest_AI ### Ownership and investors - Raised: more than $21M in total, as of July 2026 (source https://smallest.ai/blog/series-a-funding-13m-next-generation-voice-ai) - Backed by: Seligman Ventures, Sierra Ventures, 3one4 Capital, Better Capital, Upsparks Capital, Schema Ventures, DeVC, Tiny VC, Mission Street Capital ### Getting support - Discord: https://discord.com/invite/xNBrdjdBa6 - Support email: support@smallest.ai - Help: https://docs.smallest.ai/voice-agents/resources/getting-help - Help: https://status.smallest.ai/ - Sources: https://smallest.ai/presskit, https://smallest.ai/blog/series-a-funding-13m-next-generation-voice-ai, https://smallest.ai/privacy-policy, https://smallest.ai/terms-of-service, https://smallest.ai/.well-known/security.txt, https://docs.smallest.ai/api-reference/concurrency-and-limits, https://siliconangle.com/2025/10/09/exclusive-voice-ai-developer-smallest-ai-nabs-8m-investment/ ## Live (updated 2026-11-07 16:01 UTC) - Vendor status page: not read, no machine-readable status found - github `smallest-inc/smallest-python-sdk` v2.1.0, released 2025-03-04 - npm `@developer-smallestai/smallest-mcp-server` 0.11.2 - npm `smallestai` 1.0.1 - pypi `smallestai` 5.14.0, released 2026-10-07 - security.txt: valid, expires 2026-12-31T23:59:59.000Z - Always current: https://www.anchorterminal.com/api/v1/live/smallest-pulse-stt.json ## Probe metrics Not measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score. ## Prices | Item | Price | Unit | Note | | --- | --- | --- | --- | | Pulse pre-recorded | $0.003 | per minute of audio | pay as you go | | Pulse streaming | $0.004 | per minute of audio | pay as you go | | Pulse Pro pre-recorded | $0.0035 | per minute of audio | the pricing page also shows $0.004 | | Pulse 2.0 streaming | $0.005 | per minute of audio | Standard plan, per the Pulse changelog | Across all listings: https://www.anchorterminal.com/prices/index.md ## What a task costs (from the prices above) | Task | Cost | At | | --- | --- | --- | | Transcribe a one-hour meeting (60 minutes of recorded audio) | $0.18 to $0.30 | Pulse pre-recorded to Pulse 2.0 streaming | The quantity times this listing's own price, from its cheapest item in that unit to its dearest. Plans, minimums and free allowances aren't counted. ## Strengths - OpenAPI 3.1 and AsyncAPI 2.6 files cover the batch route and the streaming socket, with llms.txt and Markdown copies of every page - Public prices of $0.003 a minute for batch, $0.004 for streaming and $0.005 for Pulse 2.0, with $10 of trial credits - Dated Pulse changelog with nine entries between 30 August and 2 November 2026 - Opt-in diarisation, word timestamps, PII and PCI redaction, keyword boosting and turn detection on the same endpoint - Region-pinned hosts in Mumbai and Oregon for data residency ## Weaknesses - Health checks on the shared API failed in all three regions on 9, 10, 13, 14 and 15 September 2026, the longest about 64 minutes - Keys grant full account access with no scopes, and the docs pass keys or tokens as URL query parameters on WebSockets - Zero data retention is Enterprise only or a $1,000 a month add-on, and no retention period is published for other plans - No published SLA below Enterprise, and the terms allow changes or discontinuation without notice - Several changelog entries correct response fields the reference had documented wrongly ## Before you call it (notes for agents) 1. Pass `?model=` on every call, since a missing model returns 400; use `pulse-pro` for English files and `pulse` for other languages or streaming 2. Send booleans such as `diarize` and `word_timestamps` as the lowercase strings "true" or "false" 3. Retry 429 with backoff from 500 ms up to 30 s and honour Retry-After; keep within 25 batch requests a minute on the Standard plan 4. Use the region-pinned host api.us.smallest.ai or api.india.smallest.ai when the audio must stay in one region 5. Read speaker labels only from final events, because interim streaming events carry no `words[]` ## Connect Install: ```bash pip install smallestai ``` First request: ```bash curl -X POST "https://api.smallest.ai/waves/v1/stt/?model=pulse&language=en" \ -H "Authorization: Bearer $SMALLEST_API_KEY" -H "Content-Type: application/json" \ -d '{"url": "https://github.com/smallest-inc/cookbook/raw/main/speech-to-text/getting-started/samples/audio.wav"}' ``` Claude Code: ```bash claude mcp add smallest -- npx -y @developer-smallestai/smallest-mcp-server ``` ## Similar tools Ranked by shared capabilities, then score. Same-category tools with no shared capability key are listed last. | Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown | | --- | --- | --- | --- | --- | --- | --- | | Azure AI Speech speech-to-text | BB | 75.1 | 71 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/azure-speech-to-text.md | | OpenAI Speech to Text | BB | 75 | 74 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/openai-speech-to-text.md | | Deepgram Speech-to-Text (Nova-3, Flux) | BB | 74.4 | 86 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/deepgram-stt.md | | Amazon Transcribe | BB | 73.6 | 103 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/amazon-transcribe.md | | ElevenLabs Scribe Speech to Text API | BB | 73 | 113 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/elevenlabs-scribe.md | | Gladia Speech-to-Text API + MCP | B | 69.8 | 225 | speech.stt, speech.streaming, speech.batch, speech.diarisation, speech.languages | no | https://www.anchorterminal.com/tools/gladia-stt.md | ## Panel reviews (0) Reviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): . Desk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md ## Notable - One endpoint, `POST /waves/v1/stt/?model=pulse|pulse-pro` for files and `wss://api.smallest.ai/waves/v1/stt/live?model=pulse|pulse-2` for live audio (source: ) - Pulse 2.0 (`pulse-2`) is English-only streaming with built-in end-of-turn detection, emotion and gender on by default, at $0.005 a minute (source: ) - OpenAPI 3.1 for the HTTP routes and AsyncAPI 2.6 for the TTS, STT and speech-to-speech sockets (source: ) - Standard plan limits are 100 concurrent streaming sessions and 25 batch requests a minute per model (source: ) - Dated Pulse changelog entries from 30 August to 2 November 2026, several of them corrections that bring the docs in line with what the API sends (source: ) - Pulse and Lightning can be self-hosted on Docker or Kubernetes under an enterprise licence key (source: ) - #13 of 19 in Best speech-to-text APIs for AI agents: https://www.anchorterminal.com/best/speech-to-text/index.md - All 320 stt comparisons: https://www.anchorterminal.com/compare/speech-to-text/index.md ## Compare - [Amazon Transcribe vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/amazon-transcribe-vs-smallest-pulse-stt.md): BB 73.6 vs B 65 - [AssemblyAI Speech-to-Text (Universal) vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/assemblyai-stt-vs-smallest-pulse-stt.md): B 66.8 vs B 65 - [Azure AI Speech speech-to-text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/azure-speech-to-text-vs-smallest-pulse-stt.md): BB 75.1 vs B 65 - [Cartesia Ink vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/cartesia-ink-stt-vs-smallest-pulse-stt.md): B 67 vs B 65 - [Deepgram Speech-to-Text (Nova-3, Flux) vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/deepgram-stt-vs-smallest-pulse-stt.md): BB 74.4 vs B 65 - [ElevenLabs Scribe Speech to Text API vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/elevenlabs-scribe-vs-smallest-pulse-stt.md): BB 73 vs B 65 - [Giggy Speech to Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/giggy-stt-vs-smallest-pulse-stt.md): C 54.5 vs B 65 - [Gladia Speech-to-Text API + MCP vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/gladia-stt-vs-smallest-pulse-stt.md): B 69.8 vs B 65 - [Google Cloud Speech-to-Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/google-speech-to-text-vs-smallest-pulse-stt.md): B 66.6 vs B 65 - [Groq Speech-to-Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/groq-speech-to-text-vs-smallest-pulse-stt.md): BB 70.1 vs B 65 - [IBM Watson Speech to Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/ibm-watson-stt-vs-smallest-pulse-stt.md): C 54.5 vs B 65 - [Inworld STT vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/inworld-stt-vs-smallest-pulse-stt.md): C 54.7 vs B 65 - [MiniMax Speech-to-Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/minimax-stt-vs-smallest-pulse-stt.md): D 48.4 vs B 65 - [Mistral Voxtral Transcribe vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/mistral-voxtral-transcribe-vs-smallest-pulse-stt.md): B 63.1 vs B 65 - [OpenAI Speech to Text vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/openai-speech-to-text-vs-smallest-pulse-stt.md): BB 75 vs B 65 - [Rev AI Speech-to-Text API vs Smallest.ai Pulse STT](https://www.anchorterminal.com/compare/rev-ai-stt-vs-smallest-pulse-stt.md): C 55.5 vs B 65 - [Smallest.ai Pulse STT vs Soniox Speech-to-Text](https://www.anchorterminal.com/compare/smallest-pulse-stt-vs-soniox-stt.md): B 65 vs B 68.5 - [Smallest.ai Pulse STT vs Speechmatics Speech-to-Text](https://www.anchorterminal.com/compare/smallest-pulse-stt-vs-speechmatics-stt.md): B 65 vs B 67.8 ## Verify this listing For the vendor. The badge or a plain link to this page verifies the listing, from a page on smallest.ai or one of its subdomains, or the README at the root of github.com/smallest-inc/smallest-python-sdk. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{"slug": "smallest-pulse-stt", "url": "…"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify HTML badge: ```html Smallest.ai Pulse STT on Anchor Terminal ``` Markdown badge, for a README: ```markdown [![Smallest.ai Pulse STT on Anchor Terminal](https://www.anchorterminal.com/badges/smallest-pulse-stt.svg)](https://www.anchorterminal.com/tools/smallest-pulse-stt) ``` Plain link: ```html Smallest.ai Pulse STT on Anchor Terminal ``` ## Share this listing For the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say Smallest.ai Pulse STT is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score. - Dark: https://www.anchorterminal.com/assets/share/smallest-pulse-stt-dark.png - Light: https://www.anchorterminal.com/assets/share/smallest-pulse-stt-light.png --- Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp