API access to native Teams interpretation channels: simultaneous audio uplinks and downlinks

Question

API access to native Teams interpretation channels: simultaneous audio uplinks and downlinks

Murat KINACI 0

Does Microsoft provide a supported API, SDK, or partner integration for simultaneous bidirectional access to the native interpretation-language channels in a Teams meeting?

CONTEXT

I am developing a Microsoft Teams edition of RSI Router, a macOS application that routes live interpretation audio between online meetings and multichannel venue audio equipment.

The existing Zoom edition sends independent external interpretation-audio feeds into a meeting and receives separate interpretation-language streams simultaneously within that same meeting. I need equivalent operation in Teams, adapted to Teams' supported capacity.

This question concerns Teams' native human language-interpretation feature—not translated captions, transcripts, AI interpretation, or ordinary mixed meeting audio.

REQUIRED AUDIO FLOW

Venue → Teams: Independent external audio feeds from venue interpreters must reach their designated native Teams interpretation-language channels, rather than the main meeting mix.

Teams → venue: Multiple native interpretation-language streams must be received concurrently and routed independently to separate hardware audio channels. Original/Floor audio must also be separately identifiable.

Both directions must operate at the same time in one meeting. Attendees must retain the normal Teams interpretation-language selection experience.

QUESTIONS

Language discovery and stable identifiers

Which supported interface can enumerate the meeting's interpretation languages and provide stable identifiers for routing? How are language identifiers affected by reconnection, interpreter reassignment, or changes to language pairs?

Concurrent language reception

Can an application receive multiple interpretation-language streams simultaneously, alongside separately identifiable original/Floor audio? Is original audio blended into each received language stream, and can that blending be disabled or controlled?

Language-specific external audio transmission

Can an application publish independent external PCM audio feeds into designated native interpretation-language channels? Which methods assign or use the interpreter role and select the outgoing language? Please distinguish operations requiring organizer interaction from those available programmatically.

Simultaneous operation and capacity

Can these uplinks and downlinks operate concurrently within one meeting? What limits apply to publishers, subscribers, application instances, participant identities, and language streams?

Microsoft's language-interpretation documentation describes up to 16 language pairs. I understand this does not establish 16 independently accessible SDK audio streams plus Floor. Please clarify the actual API-level limits and any supported multi-session arrangements.

Supported implementation and access requirements

Please identify the applicable SDK/package, versions, methods, and official documentation or samples. What deployment environments, permissions, administrator consent, licensing, and approval requirements apply? The operator application is macOS-based; please clarify whether a separate Windows Server/Azure media worker would be required.

IMPORTANT DISTINCTION

I have not established that generic Microsoft Graph calling, mixed meeting audio, active-speaker streams, or AudioSocket.Send provides native interpretation-language access. Please do not treat ordinary meeting-audio transmission or reception as equivalent without identifying the language-specific mechanism.

If public interfaces do not support these operations, please confirm which requirements are unsupported and whether a supported partner integration or other official route exists. Please identify any changes that an alternative would require to the organizer, interpreter, or attendee workflow.

INITIAL VALIDATION TARGET

The initial proof would use one meeting, separately identifiable Floor and two interpretation languages, two independent external audio uplinks, and simultaneous independent language downlinks. Attendees would verify the feeds through normal Teams language selection. We would also check mapping after reconnection, original-audio blending, and unintended feedback or cross-routing.

No live Teams interpretation-channel proof has been completed. This is an API-capability and supported-architecture question, not a report of a runtime error.

Reference:

https://support.microsoft.com/en-us/teams/meetings/use-language-interpretation-in-microsoft-teams-meetings

Thank you.

0 comments

1 answer

Your answer

Answer 1

Hi @Murat KINACI

Currently ,I am not aware of any public API, SDK, or documented partner integration that provides direct access to native Teams Language Interpretation channels as separate media streams. The available documentation describes the end-user and organizer experience for Language Interpretation, but does not document APIs for:

Enumerating interpretation-language channels.

Obtaining stable identifiers for interpretation streams.

Subscribing to multiple interpretation-language streams concurrently.

Receiving Floor audio separately from interpretation audio.

Publishing audio directly into a designated interpretation-language channel.

Programmatically assigning interpreter routing for media streams.

Additionally, from the reference you mentioned, Microsoft's documentation confirms that meeting recordings only capture the original/main audio channel, not interpreter audio, reinforcing that interpretation audio isn't treated as an independently accessible stream even in already-supported scenarios like recording.

User's image

Reference: Use language interpretation in Microsoft Teams meetings

Please note that this summary is based on my own findings and may not fully address your concerns. To help you reach your goal more effectively, I recommend engaging with [GitHub Community Forum] for a deeper technical dive or to connect with individuals who have relevant experience and expertise. Some approaches may behave differently or be restricted depending on your specific environment and configuration. These forums include many experienced developers and Microsoft specialists who can assist with troubleshooting and guidance.

I hope this helps.

Murat KINACI 0 Reputation points

2026-09-11T20:08:26.7266667+00:00

Hello Hin Vo,

Thank you for taking the time to investigate this and provide such a detailed response. I appreciate the documentation references and your careful distinction between the end-user interpretation feature and the capabilities exposed through documented developer interfaces.

To clarify my immediate requirement, I only need to receive multiple interpretation-language feeds simultaneously from the same Teams meeting and route each to a separate audio output in a conference venue. I do not need to publish audio into Teams interpretation channels or programmatically assign interpreter roles.

Importantly, a small amount of original/Floor audio in the background of each interpretation feed is acceptable. Completely isolated interpretation-only audio is not essential, provided that each selected language can be received and routed independently.

Your findings are helpful in narrowing the investigation. If you know of a supported approach for this more limited, receive-only use case, I would greatly appreciate any further guidance.

Thank you again for your time and assistance.

Kind regards, Murat KINACI Idil Software — RSI Router