MeetStream partners: in-house transcription and model providers
MeetStream transcribes meetings in-house, and Deepgram, AssemblyAI, Sarvam and JigsawStack run on your own keys. Voice agents add the reasoning, voice and avatar providers you name on a request.
How a provider is chosen
Name the provider when you create the bot. One runs per request. The next meeting can name another one.
In-house transcription, plus your own keys
MeetStream transcribes after the call or live during it, for $0.10 per bot-hour. Deepgram, AssemblyAI, Sarvam and JigsawStack also run on your own provider keys, and the platform captions need no key.
Post-call or live, for $0.10 per bot-hour and no third-party key. Post-call runs spot the language on their own and translate when you ask. Live runs push raw, word or sentence chunks to your webhook.
Post-call transcripts on nova-3, with diarisation, punctuation and utterances on by default. Deepgram Streaming pushes raw, word or sentence chunks live. Runs on your own Deepgram key.
Post-call transcripts on universal-2, with speaker labels. PII redaction and chapters are options. AssemblyAI Streaming sends chunks while the call runs. Runs on your own AssemblyAI key.
Post-call transcripts on saaras:v3, built for Indic languages and English code-switching. Transcribe and translate are both modes you can name on the request. Runs on your own Sarvam key.
Post-call transcripts with auto-detect, or any of 162 languages named on the request. Output arrives split per speaker. Runs on your own JigsawStack key.
Native Google Meet and Teams captions, sent live with nothing to set up. Pair it with a post-call run when you also want a stored transcript.
The transcription integration holds the options table, and the transcription API covers the request itself.
Model, voice and avatar providers
A voice agent is a saved config. These are the providers you can name inside it.
Realtime models carry a whole voice-to-voice turn. In pipeline mode OpenAI is a reasoning choice. Whisper is a speech-to-text choice, and OpenAI voices read the reply out loud.
A reasoning choice in pipeline mode. It picks the reply between the speech engine and the voice.
Gemini native audio runs a realtime turn end to end. Google is also a reasoning choice in pipeline mode.
Named at every stage. It runs realtime turns, pipeline speech to text, pipeline reasoning, and pipeline voices.
A text-to-speech choice in pipeline mode, so the agent answers in the voice you pick.
Virtual avatars. Anam gives the bot a lip-synced face, set on the saved agent beside the voice.
The voice agents page sets out realtime and pipeline mode side by side.
Where each provider runs
One row per provider, and the key it runs on.
| Provider | Where it runs on MeetStream | Key |
|---|---|---|
| MeetStream in-house | Post-call and live transcripts, language detection, translation | None needed |
| Deepgram | Post-call transcripts, live chunks, pipeline speech to text | Yours |
| AssemblyAI | Post-call transcripts, live chunks, pipeline speech to text | Yours |
| Sarvam | Post-call transcripts, Indic languages, translate mode | Yours |
| JigsawStack | Post-call transcripts, auto-detect across 162 languages | Yours |
| Meeting captions | Live captions from Google Meet and Microsoft Teams | None needed |
| OpenAI | Realtime turns, pipeline reasoning, Whisper speech to text, voices | Yours |
| Anthropic | Pipeline reasoning | Yours |
| Gemini native audio realtime, pipeline reasoning | Yours | |
| xAI | Realtime turns, pipeline speech to text, reasoning and voices | Yours |
| ElevenLabs | Pipeline text to speech | Yours |
| Anam | Virtual avatars on a saved agent | Yours |
Your keys, your accounts
Connect a provider once in the dashboard. The engine then runs on your account. Keys stay there and never travel in a request.
Rather run your own model? Take raw frames from the real-time audio socket and do the rest yourself.
One request, whichever provider you pick.
Speaker labels come from the stream, so they hold whichever option runs.
Got a question? We got the answer.
Common questions about which providers are available, whose key runs them, and how to swap.
Which transcription engines can I use?
Start with MeetStream's in-house transcription, after the call or live during it, for $0.10 per bot-hour and no provider key. Deepgram, AssemblyAI, Sarvam and JigsawStack are also available with your own provider keys, and the meeting platform's own captions need no key.
Which model providers can a voice agent use?
Realtime mode runs OpenAI Realtime, Google Gemini native audio or xAI. Pipeline mode splits the turn in three. Speech to text takes Deepgram, AssemblyAI, OpenAI Whisper or xAI. Reasoning takes OpenAI, Anthropic, Google or xAI. The voice takes OpenAI, ElevenLabs or xAI.
Do I need my own provider key?
No. In-house transcription and platform captions need no third-party key. To run Deepgram, AssemblyAI, Sarvam or JigsawStack, save that provider's key in the dashboard. Transcription then runs on your account and the $0.10 add-on is waived.
Can I switch provider between meetings?
Yes. One option runs per request, and the next meeting can name a different one. Re-transcribe a stored recording later, in another language or on your own provider key.
Do speaker labels change with the provider?
Names come from the stream each person is on, plus the speaker timeline. Switch engines and the words change while the names stay put.
Can I run my own model instead?
Yes. Subscribe to the real-time audio WebSocket. You get 48 kHz mono PCM16 frames straight from the room. Each frame carries the speaker ID and name, about 200 ms behind the talk.
How do I check transcript quality?
Run it on your own meetings with the $5 free credit. Re-transcription runs a stored recording again in another language or on your own provider key, so you can compare outputs on the audio you care about.
Try in-house transcription on your next meeting
$5 of free credit, live and post-call transcription, and no provider key needed.