Configuration

Basic Configuration

Edit /etc/deepspeech/config.yaml:

yaml

Copy

model:
  path: /var/lib/deepspeech/model
  beam_width: 500
  lm_alpha: 0.75
  lm_beta: 1.85
audio:
  sample_rate: 16000
  vad_aggressiveness: 3
server:
  host: 0.0.0.0
  port: 8000
  max_concurrent: 4
gpu:
  enabled: true
  device: 0
Performance Settings

yaml Copy

inference:
  batch_size: 16
  num_threads: 4
streaming:
  chunk_size: 1024
  overlap: 160