NagentArtefacts
Dev & Technology
Enterprise and White Glove
Company
Gladia connects to a Nagent workspace with an API key. Once it is connected, agents can call 10 Gladia actions, such as "Get live transcription result", "Get Pre-recorded Job" and "Get Transcription Audio File (Deprecated)". Nothing is enabled on connect: each action is allowed one at a time, and an action with side effects runs or waits for a person according to the agent's level.
Every operation an agent can call against Gladia, with input parameters and output schema.
GLADIA_DELETE_LIVE_SESSIONTool to delete a live transcription session and all its associated data (audio file, transcription). Use when permanently removing a live session from Gladia.
Input parameters
UUID of the live transcription session to delete
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_DELETE_PRE_RECORDED_JOBTool to delete a pre-recorded transcription job and all its associated data (audio file, transcription). Use when permanently removing a transcription job from Gladia.
Input parameters
UUID of the pre-recorded transcription job to delete
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_GET_LIVE_TRANSCRIPTION_RESULTTool to retrieve metadata and results of a live transcription job. Use when you need detailed status or results for a specific live transcription session.
Input parameters
Unique identifier of the live transcription job
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_GET_PRE_RECORDED_JOBTool to retrieve a pre-recorded transcription job's metadata, status, and results by ID. Use when checking the status or retrieving results of a specific pre-recorded job.
Input parameters
Id of the pre recorded job
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_GET_TRANSCRIPTION_AUDIO_FILE_DEPRECATEDDEPRECATED - use GET /v2/pre-recorded/{id}/file instead. Download the audio file used for a transcription job. Use when you need to retrieve the original audio file associated with a specific transcription job ID.
Input parameters
Id of the transcription job
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_INITIATE_LIVE_SESSIONInitiates a live (real-time) transcription session with Gladia's API. Returns a WebSocket URL that you can connect to for streaming audio data in real-time. The WebSocket URL includes an embedded authentication token and should be used promptly. After initiating, connect to the returned WebSocket URL and send audio chunks. Use GET /v2/live/{id} to retrieve the full transcription results after the session ends.
Input parameters
Transcription model to use. 'solaria-1' is the default model. 'solaria-2' may require a higher subscription plan.
Processing region for the session. Use 'us-west' for US West or 'eu-west' for EU West. If not specified, the default region is used.
Enable callback notifications. When true, configure callback_config to specify the callback URL.
Number of audio channels. Use 1 for mono (most speech), 2 for stereo. Maximum 8 channels supported.
Audio encoding format for the stream. Use 'wav/pcm' for raw PCM, 'wav/alaw' for A-law, or 'wav/ulaw' for µ-law encoding.
Audio sample bit depth. Common values: 8-bit for telephony, 16-bit for standard audio, 24/32-bit for high-quality audio.
Duration of silence in seconds that triggers end of utterance. Lower values (0.01-0.1) for responsive transcription, higher values (0.5-2.0) for longer pauses.
Audio sample rate in Hz. Use 8000 for telephony, 16000 for speech, 44100/48000 for high-fidelity audio.
Pre-processing settings for audio enhancement and speech activity detection thresholds.
Callback delivery configuration. Required if callback is true. Specify callback URL and which events to send.
Custom key/value metadata to attach to the session for tracking or identification purposes.
Language configuration object. Example: {'languages': \['en'\]} for English, or enable code-switching for multilingual audio.
Configuration for which message types to receive over WebSocket (transcripts, partial results, etc.).
Post-processing settings for summarization and chapterization of the transcript after the session ends.
Realtime processing options including custom vocabulary, translation, named entity recognition (NER), and sentiment analysis.
Maximum duration in seconds before forcing an utterance end, even without detecting silence. Range: 5-60 seconds.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_INITIATE_PRE_RECORDED_TRANSCRIPTIONTool to initiate a pre-recorded transcription job. Use when you have an audio URL and need asynchronous transcription results.
Input parameters
Transcription model. 'solaria-1' (default) or 'solaria-2' (may require higher tier plan).
Enable webhook callback when transcription completes.
URL to an audio or video file. Can be a public URL or a Gladia file URL from the upload endpoint (format: https://api.gladia.io/file/<id>).
Enable sentence segmentation for better readability.
Enable subtitle generation. Set to true to generate SRT/VTT files.
Enable speaker diarization to identify who said what. Set to true to detect speakers.
Enable translation of transcription. Set to true to translate to other languages.
Enable summarization of transcription. Set to true to get a summary.
Configuration for webhook callback when transcription completes.
Arbitrary metadata to attach to the job for tracking and organization.
Configuration for preferred language handling.
Configuration for subtitle generation.
Configuration for speaker diarization.
Configuration for translation of transcription.
Enable enhanced punctuation and casing for cleaner transcripts.
Configuration for summarization of transcription.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_LIST_LIVE_TRANSCRIPTIONSTool to list live transcription jobs. Use when you need an overview of live transcription sessions with optional filtering and pagination. Use after setting up live transcription.
Input parameters
Filter items relevant to a specific date in ISO format (YYYY-MM-DD).
Max number of items to return.
The starting point for pagination; 0 starts from the first item.
Filter by item status; accepts multiple values. Valid statuses: 'queued', 'processing', 'done', 'error'.
Filter for items after the specified ISO date-time (use with before_date for a range).
Include items that occurred before the specified ISO date-time.
Filter by custom metadata object (e.g., {'user': 'John Doe'}).
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_LIST_PRE_RECORDED_JOBSTool to list all pre-recorded transcription jobs matching the query parameters. Use when you need to retrieve a paginated list of transcription jobs with optional filtering by date, status, or custom metadata.
Input parameters
Filter items relevant to a specific date in ISO format (YYYY-MM-DD).
The maximum number of items to return. Useful for pagination and controlling data payload size.
The starting point for pagination. A value of 0 starts from the first item.
Filter the list based on item status. Accepts multiple values from the predefined list.
Filter for items after the specified date. Use with `before_date` for a range. Date in ISO format.
Include items that occurred before the specified date in ISO format.
Custom metadata filter to match jobs, e.g., {"user": "John Doe"}.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GLADIA_UPLOAD_AUDIO_VIDEO_FILEUpload an audio or video file to Gladia for speech-to-text transcription. Returns a URL that can be used with the pre-recorded transcription endpoint. Supports common audio formats (WAV, MP3, FLAC, OGG) and video formats (MP4, MKV, etc.).
Input parameters
The audio or video file to upload. Supports common formats like WAV, MP3, FLAC, MP4, etc.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
Agents can call 10 Gladia actions on Nagent, including "Get live transcription result", "Get Pre-recorded Job" and "Get Transcription Audio File (Deprecated)". Get live transcription result: Retrieve metadata and results of a live transcription job. Each action is listed on this page with its input parameters and its output.
Gladia connects with an API key, under your workspace's own connection. Nothing is enabled on connect: each action is allowed one at a time and can be scoped to the agents that need it.
Get live transcription result takes 1 required input: id. It returns data, error and successful.