​
Overview

GladiaSTTService provides real-time speech recognition using Gladia’s WebSocket API with support for 99+ languages, custom vocabulary, translation, sentiment analysis, and advanced audio processing features.

API Reference

Complete API documentation and method details

Gladia Docs

Official Gladia documentation and features

Example Code

Working example with interruption handling

​
Installation

To use Gladia services, install the required dependency: 
pip install "pipecat-ai[gladia]"
You’ll also need to set up your Gladia API key as an environment variable: GLADIA_API_KEY.
Get your API key from Gladia.

​
Frames

​
Input

  • InputAudioRawFrame - Raw PCM audio data (16-bit, 16kHz, mono)
  • STTUpdateSettingsFrame - Runtime transcription configuration updates
  • STTMuteFrame - Mute audio input for transcription

​
Output

  • InterimTranscriptionFrame - Real-time transcription updates
  • TranscriptionFrame - Final transcription results
  • TranslationFrame - Real-time translation results (when enabled)
  • ErrorFrame - Connection or processing errors

​
Models

Gladia offers several models optimized for different use cases:
ModelDescriptionBest For
solaria-1Latest general-purpose modelHigh accuracy, balanced performance
See Gladia’s model documentation for detailed comparisons.

​
Language Support

Gladia STT supports 99+ languages with automatic detection and code-switching:
Common languages:
  • Language.EN - English - en
  • Language.ES - Spanish - es
  • Language.FR - French - fr
  • Language.DE - German - de
  • Language.IT - Italian - it
  • Language.JA - Japanese - ja

​
Advanced Features

​
Automatic Language Detection

  • Single language: Fixed language for entire session
  • Multiple languages: Auto-detect per utterance with code_switching=True
  • No languages specified: Auto-detect from all supported languages

​
Custom Vocabulary

  • Add domain-specific terms with bias intensity (0.0-1.0)
  • Mix strings and CustomVocabularyItem objects
  • Configure default intensity for simple strings

​
Real-time Translation

  • Translate to multiple target languages simultaneously
  • Enhanced model for higher accuracy
  • Align translations with original utterances

​
Usage Example

​
Basic Configuration

Initialize the GladiaSTTService and use it in a pipeline: 
from pipecat.services.gladia.stt import GladiaSTTService
from pipecat.services.gladia.config import GladiaInputParams, LanguageConfig

# Simple setup
stt = GladiaSTTService(
    api_key=os.getenv("GLADIA_API_KEY"),
)

# Use in pipeline
pipeline = Pipeline([
    transport.input(),
    stt,
    context_aggregator.user(),
    llm,
    tts,
    transport.output(),
    context_aggregator.assistant()
])

​
Dynamic Configuration

Make settings updates by pushing an STTUpdateSettingsFrame for the GladiaSTTService: 
from pipecat.frames.frames import STTUpdateSettingsFrame

await task.queue_frame(STTUpdateSettingsFrame(
    language=Language.FR,
))

​
Multi-language Configuration

from pipecat.services.gladia.config import LanguageConfig

# Multi-language with auto-detection
params = GladiaInputParams(
    language_config=LanguageConfig(
        languages=["en", "es", "fr"],  # English, Spanish, French
        code_switching=True  # Auto-detect language changes
    )
)

stt = GladiaSTTService(
    api_key=os.getenv("GLADIA_API_KEY"),
    params=params
)

​
Real-time Translation

from pipecat.services.gladia.config import (
    RealtimeProcessingConfig,
    TranslationConfig
)

# Enable real-time translation
params = GladiaInputParams(
    language_config=LanguageConfig(languages=["en"]),
    realtime_processing=RealtimeProcessingConfig(
        translation=True,
        translation_config=TranslationConfig(
            target_languages=["es"],
            match_original_utterances=True
        )
    )
)

stt = GladiaSTTService(
    api_key=os.getenv("GLADIA_API_KEY"),
    params=params
)

​
Custom Vocabulary

from pipecat.services.gladia.config import (
    CustomVocabularyConfig,
    CustomVocabularyItem,
    RealtimeProcessingConfig
)

# Add domain-specific vocabulary
custom_vocab = CustomVocabularyConfig(
    vocabulary=[
        CustomVocabularyItem(value="Pipecat", intensity=0.9),
        CustomVocabularyItem(value="WebRTC", intensity=0.8),
        "JavaScript",  # Simple string with default intensity
    ],
    default_intensity=0.6
)

params = GladiaInputParams(
    realtime_processing=RealtimeProcessingConfig(
        custom_vocabulary=True,
        custom_vocabulary_config=custom_vocab
    )
)

​
Metrics

The service provides comprehensive metrics:
  • Time to First Byte (TTFB) - Latency from audio input to first transcription
  • Processing Duration - Total time spent processing audio
Learn how to enable Metrics in your Pipeline.