IBM Watson Speech to Text vs Inworld STT

IBM Watson Speech to Text 40

IBM Cloud has a status page at cloud.ibm.com/status, but it is drawn by script and showed no components or history to our reader, read as gemini-live's script-shell page was (10 of 20, and 5 for no readable history). Plus allows up to 100 concurrent transcription requests across WebSocket and HTTP (15). No 429 or backoff guidance was found in the API reference or the docs repository (0). No SLA was read for the service (0, unchecked). Next-generation and large speech models are GA, language by language (10). 10 + 5 + 15 + 0 + 0 + 10 = 40.

Inworld STT 60

Hosted reading of the checklist. status.inworld.ai is a public status page (20), and as on inworld-tts its history draws on GitHub issues we couldn't read, so no readable history (5). Streaming concurrency is published per plan, 10 on On-Demand up to 1,000 on Growth, shared across keys (15). 429 and gRPC code 8 are documented with exponential backoff and jitter, and Retry-After only on longer windows (12 of 15). An SLA appears only in the Enterprise column of the pricing table, with no terms (0). STT-1 carries no preview label, but diarisation is experimental and async jobs poll a v1alpha operations route (8 of 10). 20 + 5 + 15 + 12 + 0 + 8 = 60.

IBM Watson Speech to Text 80

Closed service under the IBM Cloud terms (15). The data security page states default logging for improvement with an opt-out, async retention, GDPR deletion by customer ID and plan-level separation, and they agree with each other. The DPA wasn't read (20 of 30). Previous-generation models carried a dated deprecation of 15 March 2022 and removal on 31 July 2023, and release notes say changes are compatible unless noted (18 of 20). Ten locations are named, with sub-processors not read (12 of 20). Total 65.

Inworld STT 60

Closed service with clear terms naming Theai, Inc. dba Inworld AI (15). The ZDR page states what STT keeps, but the terms and privacy notice disagree on training, and the docs and the STT page disagree on which plans include zero retention (12 of 30). No dated STT deprecation or change notices are public (8 of 20). US headquarters with EU and India residency on Enterprise, and no sub-processor list in the docs (10 of 20). Total 45.

IBM Watson Speech to Text 66

API reading of the checklist. Timestamps, word confidence, alternatives and speaker labels are opt-in (15 of 25). Async jobs and custom models list, and async results can go to a callback URL (12 of 20). Structured errors and warnings an agent can act on (15 of 20). Async jobs take a user_token, results have a TTL, and jobs can be deleted, with no idempotency key (12 of 20). Most parameters are optional. Official SDKs including Python, Node and Swift, though the Python and Node SDKs last released in January 2026 (12 of 15). Total 66.

Inworld STT 56

API reading of the checklist. Voice Profile, word timestamps and diarisation are opt-in, and responses carry a usage block (18 of 25). No lists to page. Async jobs return an operation to poll (8 of 20). Errors carry a numeric gRPC code and message on both endpoints, but a malformed first message closes the socket with 1005 and no frame (12 of 20). Sync calls are stateless and async calls return an operation, with no idempotency key, and empty transcripts and add-ons are billed (10 of 20). A call needs a model, an encoding and the audio. No STT SDK was found, only the CLI and its MCP server and REST examples (8 of 15). Total 56.

IBM Watson Speech to Text 33

Read as a model. The newest release note is 29 June 2026 (parameter_set for large speech models), 130 days ago (10). No dated entries in the 90 days from 8 August (0). Release notes, IBM support and a public docs repository last changed 30 June 2026 (8 of 15). Official SDKs, ibm-watson 11.2.0 on PyPI (26 January 2026) and 12.2.0 on npm (January 2026) (10 of 15). CI not read (5 of 10). Total 33.

Inworld STT 41

Read as a model. No public STT changelog. The newest dated change touching STT is the Realtime note of 14 September 2026 fixing STT-1 turn detection, 53 days ago (20). The CLI, which carries the STT command and MCP tool, shipped 1.4.0 to 1.5.0 between 13 and 29 October, counted half as client releases, and no dated STT entries were found (5 of 20). Discord and support@inworld.ai, with no evidence of answers read (6 of 15). No official STT SDK; the CLI is current (5 of 15). CLI on npm under MIT, current, CI not read (5 of 10). Total 41.

IBM Watson Speech to Text 68

Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. IAM API keys per instance with Reader, Writer and Manager roles and short-lived bearer tokens (30). The WebSocket interface takes the token as an access_token query parameter, so 10 off (20 of 30). Request and response data are logged by default to improve the service, and any account can opt out per account or per request (10 of 20). Async results expire after a week by default (results_ttl shortens it), a delete method exists, and DELETE /v1/user_data removes data by customer ID (10 of 15). Activity Tracker events record changes to the service's state (10 of 15). Valid security.txt to 6 December 2026, PSIRT and HackerOne, HIPAA readiness on Premium. Certifications weren't read (18 of 20). Total 68.

Inworld STT 63

Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Workspace API keys sent as Basic credentials, deleted to revoke, with separate keys per environment advised. The CLI mints read-only keys unless --write is given, but write toggles cover only the Voice and Router APIs, so no key is limited to one product. Browsers use short-lived tokens minted server side, via the bearer_ WebSocket subprotocol, not the URL (25 of 30). The Terms (11 June 2025) say Inworld won't train generally available models on customers' non-public Materials, except feedback and flagged content, while the Privacy Notice (18 January 2026) lists training AI models among its uses, per the inworld-tts check of 4 November (12 of 20). Zero data retention keeps STT audio and transcripts in memory and out of logs, included on Developer ($300 a month) and above per the docs, while the STT page says Enterprise or a Growth add-on. Async results and audio go after 7 days, and no default period is given for logs (8 of 15). Usage is attributed per key and a Performance page shows errors and throttling (8 of 15). SOC 2 Type II and HIPAA claimed and a vulnerability report form, with no security.txt or bug bounty, per the 4 November reading (10 of 20). Total 63.