STT
Whisper-class transcription
The whisper canonical maps to Voxtral (EU default), Whisper large-v3 (Scaleway, EU) and Azure Speech (CH / EU / US). Returns json, text, srt, vtt or verbose_json.
Capability · Speech
Transcription and speech, residency-separate.
Speech-to-text and text-to-speech on the OpenAI audio shape — Swiss, EU and US only, each pinned to a same-jurisdiction provider with no cross-border failover.
POST /v1/audio/transcriptions · /v1/audio/speech
What you get
Built for regulated workloads
Transcription uses the standard /v1/audio/transcriptions multipart shape (json, text, srt, vtt, verbose_json); speech uses /v1/audio/speech. Both are deliberately Swiss + EU + US only and residency-separate — a CH request resolves to a Swiss-jurisdiction host and never fails over across a border. Anything outside those three regions returns no_route by design.
TTS
Neural text-to-speech
Voxtral (EU) and Azure neural voices (EU / US) synthesise MP3, WAV, OPUS, AAC, FLAC or PCM from the OpenAI speech shape.
Residency
No cross-border failover
STT/TTS are pinned within a jurisdiction. A Swiss call will not silently complete on an EU host — if the resident provider is unavailable, the call fails in-region rather than crossing a border.
PHI
PHI on verified hosts
PHI speech runs on Azure (CH / EU / US) and Scaleway (EU), all DPA/BAA-verified. CH text-to-speech has no Swiss PHI host today and resolves to no_route by design.
Availability & pricing
Where it runs, what it costs
Every route is region-resident and the PHI gate is enforced per call. Prices include the flat +10% gateway margin and mirror the live /v1/pricing rate card.
| Region | Provider | Model | Tier | Price |
|---|---|---|---|---|
| EU | Mistral (Voxtral) | STT · whisper | General | $0.0033 / min |
| EU | Scaleway | STT · whisper large-v3 | PHI | $0.0036 / min |
| CH | Azure Speech | STT | PHI | $0.0183 / min |
| US | Azure Speech | STT | PHI | $0.0183 / min |
| EU | Mistral (Voxtral) | TTS | General | $17.60 / 1M chars |
| EU / US | Azure Speech | TTS · neural | PHI | $16.50 / 1M chars |
| CH | — | TTS | No route | No Swiss host |
STT billed per minute of audio; TTS per character. Gateway margin included. CH STT runs on Azure today (Infomaniak Whisper is staged, async-batch).
Try it
A real call, end to end
Transcribe a consult recording in the EU with a residency-pinned Whisper.
curl
curl https://phi-cloud.com/api/v1/audio/transcriptions \
-H "Authorization: Bearer $PHI_API_KEY" \
-H "X-Region: EU" \
-H "X-PHI: true" \
-F file=@consult.mp3 \
-F model=whisper \
-F response_format=verbose_jsonresponse
x-phi-routed: scaleway-whisper-v3/EU/phi
x-phi-tier: phi
x-phi-usage: verified
{ "text": "Patient reports...", "duration": 184.2 }Residency & coverage
- CH speech-to-text → Azure Speech (Switzerland). CH text-to-speech → no_route (no Swiss PHI host yet).
- EU → Voxtral / Whisper / Azure for STT; Voxtral / Azure for TTS.
- US → Azure for both STT and TTS.
- Every other region → no_route. Speech is intentionally CH/EU/US only.
FAQ
Common questions
There is no Swiss-resident, PHI-eligible TTS host today. Rather than route Swiss PHI audio across a border, the gateway refuses it. Swiss speech-to-text is available on Azure; TTS will light up when a compliant Swiss host is verified.
No. Speech is residency-separate with no cross-border failover — that is the point. A failed CH call stays a CH error; it never completes in another jurisdiction.
STT accepts standard audio uploads and returns json, text, srt, vtt or verbose_json. TTS emits MP3, WAV, OPUS, AAC, FLAC or PCM via the OpenAI /audio/speech shape.
Ready when you are
Put speech (stt / tts) in production — without giving up your data.
Spin up a key in minutes. The residency and PHI posture above arrives unchanged.
Free to test · Prepaid credits, no subscription · No data retained