SortformerEmotionTranscription
Python template · sinapsis_diarization.templates.emotion_diarization_templates
Overview
Template that runs a pipeline for ASR with Diarization and Emotion Recognition using Parakeet as ASR, Sortformer as Diarization engine and Speechbrain as Emotion Engine.
Supported models: Parakeet -'nvidia/parakeet-tdt-0.6b-v2' -'nvidia/parakeet-tdt-0.6b-v3' Sortformer -'nvidia/diar_streaming_sortformer_4spk-v2.1'
Attributes: root_dir(str): Path to root directory. Optional audio_file_path(str): Path to audio file. Optional asr_model_name (str): Name of the ASR pipeline to use diarization_model_name (str): Name of the diarization pipeline device (str): device to run the model sample_rate (int): sample rate of the audio chunk_size_in_secs (int): Size of the chunks to divide the audio. -1 for full audio search_window (float): Size of the search window num_speakers (int): Number of speakers to predict min_emotion_confidence (float): Minimum confidence threshold to include an emotion. Default 0.5
Example agent:
agent: name: my_test_agent templates: - template_name: InputTemplate class_name: InputTemplate attributes: {} - template_name: SortformerEmotionTranscription class_name: SortformerEmotionTranscription template_input: InputTemplate attributes: /root/.cache/sinapsis audio_file_path: null asr_model_name: nvidia/parakeet-tdt-0.6b-v2 diarization_model_name: nvidia/diar_streaming_sortformer_4spk-v2.1 device: cuda sample_rate: 16000 chunk_size_in_secs: -1 search_window: 5.0
Capabilities
Capabilities: not setInstallation
Install the published package using pip or uv.
pip install sinapsis-diarization==1.0.0uv pip install sinapsis-diarization==1.0.0uv add sinapsis-diarization==1.0.0Configuration
Effective fields from the inspector schema, including nested Pydantic model fields.
Optional
Default: "cuda"
Optional
Default: null
Optional
Default: "/home/ariel/.cache/sinapsis"
Optional
Default: 16000
Optional
Default: 5
Optional
Default: "nvidia/parakeet-tdt-0.6b-v2"
Optional
Default: null
Optional
Default: -1
Maximum audio duration in seconds per emotion chunk. Default 60.0
Optional
Default: 60
Optional
Default: "nvidia/diar_streaming_sortformer_4spk-v2.1"
Minimum confidence threshold to include an emotion
Optional
Default: 0.5
Configuration example
Generated from documented defaults. Replace any placeholders and complete required nested settings before running.
templates:
- template_name: "SortformerEmotionTranscription"
class_name: "SortformerEmotionTranscription"
attributes:
device: "cuda"
root_dir: "/home/ariel/.cache/sinapsis"
sample_rate: 16000
search_window: 5
asr_model_name: "nvidia/parakeet-tdt-0.6b-v2"
audio_file_path: null
chunk_size_in_secs: -1
emotion_chunk_secs: 60
diarization_model_name: "nvidia/diar_streaming_sortformer_4spk-v2.1"
min_emotion_confidence: 0.5
Inheritance
sinapsis_diarization.templates.emotion_diarization_templates.EmotionTranscriptionBaseTemplateEnvironment variables
Environment definitions belong to their canonical Python modules.
No directly related environment variables listed.