A hosted OpenAI-compatible speech endpoint for any app
Many apps have a setting called OpenAI-compatible base URL, API URL or custom endpoint for speech. These are the values to put there.
1. Get a key
Get an access key on the pricing page: $1 once (1,000 credits) or $3 a month (3,000 credits, about 60 hours). After paying you get a page that shows the key (it starts with adt_). That key is the API key in every app below.
2. The values
| Base URL (if the field wants the base) | https://glyphcall.com/v1 |
|---|---|
| Speech endpoint (if it wants the full path) | https://glyphcall.com/v1/audio/speech |
| Transcription endpoint (full path) | https://glyphcall.com/v1/audio/transcriptions |
| API key | adt_… (sent as Authorization: Bearer) |
| TTS model | tts-1 (any value is accepted) |
| STT model | whisper-1 (any value is accepted) |
| Voice | nova, alloy, onyx… or a name from the voice list |
How to tell which URL a field wants: if its default is https://api.openai.com/v1, give the base; if the default ends in /audio/speech or /audio/transcriptions, give the full path.
3. What is supported
- Text to speech:
inputup to 2,000 characters,response_formatmp3 only,speed1 only. English by default; Spanish withlanguage: "es"and a Spanish voice. - Speech to text: a multipart
fileup to 25 MB (over 5 MB: MP3 or WAV, up to 60 minutes);response_formatjson, text, srt, vtt or verbose_json; optionallanguage. - Browser apps can call it directly: CORS allows the Authorization header.
4. Test it
curl https://glyphcall.com/v1/audio/speech \
-H "Authorization: Bearer YOUR_KEY" -H "Content-Type: application/json" \
-d '{"model":"tts-1","input":"Hello from Glyphcall Voice.","voice":"nova"}' -o hello.mp3
curl https://glyphcall.com/v1/audio/transcriptions \
-H "Authorization: Bearer YOUR_KEY" \
-F file=@clip.mp3 -F model=whisper-1
Apps people use it with: SillyTavern (TTS provider for OpenAI-compatible services; add the voice names by hand), dictation apps that accept any OpenAI-compatible endpoint, browser read-aloud extensions with an API URL field, and the OpenAI SDKs (set base_url).
Test the key first
If something does not work in the app, check the key and the endpoint with one call from a terminal:
curl https://glyphcall.com/v1/audio/speech \
-H "Authorization: Bearer YOUR_KEY" -H "Content-Type: application/json" \
-d '{"model":"tts-1","input":"Hello from Glyphcall Voice.","voice":"nova"}' -o hello.mp3
What it costs
| Call | Price |
|---|---|
| Text to speech, up to 300 characters | $0.01 (10 credits) |
| Text to speech, 301 to 2,000 characters | $0.08 (80 credits) |
| Speech to text, a file up to 5 MB | $0.01 (10 credits) |
| Speech to text, a larger MP3 or WAV up to 60 minutes / 25 MB | $0.05 (50 credits) |
A failed call is never charged. No account, no subscription unless you choose the monthly plan.
Questions
Is it really OpenAI-compatible? The request and answer shapes are OpenAI's: tested with the official OpenAI SDKs for Node and Python. The voices are Deepgram Aura-2 and the transcription model is Whisper large-v3-turbo.
What does a failed call cost? Nothing. Refused and failed calls are never charged.
Other guides: Open WebUI · LibreChat · AnythingLLM · Home Assistant · Obsidian Whisper · OpenAI tts-1 shutdown · OpenAI whisper-1 shutdown · Voice list