Implementation:Elevenlabs Elevenlabs python AudioWithTimestampsAndVoiceSegmentsResponseModel

Field	Value
source	Elevenlabs_Elevenlabs_python
domains	Audio, Text-to-Speech, Timestamps, Voice Segments
last_updated	2026-02-15

Overview

Description

AudioWithTimestampsAndVoiceSegmentsResponseModel is a Pydantic model representing an audio response that includes both character-level timestamp alignments and voice segment information. It carries base64-encoded audio data along with alignment data for both the original and normalized text, and a list of voice segments identifying which voice produced each portion of the audio. This model is auto-generated by Fern from the ElevenLabs API definition and extends UncheckedBaseModel.

Usage

This model is returned by the ElevenLabs text-to-speech API when requesting audio with timestamps and voice segments. It is useful for applications that need precise timing information for lip-syncing, subtitle generation, or voice attribution in multi-speaker scenarios.

Code Reference

Source Location

src/elevenlabs/types/audio_with_timestamps_and_voice_segments_response_model.py

Class Signature

class AudioWithTimestampsAndVoiceSegmentsResponseModel(UncheckedBaseModel):
    ...

Import Statement

from elevenlabs.types import AudioWithTimestampsAndVoiceSegmentsResponseModel

I/O Contract

Field	Type	Required	Description
audio_base_64	str	Yes	Base64 encoded audio data. Serialized with alias audio_base64.
alignment	Optional[CharacterAlignmentResponseModel]	No	Timestamp information for each character in the original text.
normalized_alignment	Optional[CharacterAlignmentResponseModel]	No	Timestamp information for each character in the normalized text.
voice_segments	List[VoiceSegment]	Yes	Voice segments for the audio.

Usage Examples

from elevenlabs.types import AudioWithTimestampsAndVoiceSegmentsResponseModel
import base64

# Typically received as part of an API response
response = AudioWithTimestampsAndVoiceSegmentsResponseModel(
    audio_base_64="SGVsbG8gV29ybGQ=",
    alignment=CharacterAlignmentResponseModel(...),
    normalized_alignment=CharacterAlignmentResponseModel(...),
    voice_segments=[VoiceSegment(...)],
)

# Decode the audio data
raw_audio = base64.b64decode(response.audio_base_64)

# Iterate over voice segments
for segment in response.voice_segments:
    print(segment)

Related Pages

• Implementation:Elevenlabs_Elevenlabs_python_StreamingAudioChunkWithTimestampsAndVoiceSegmentsResponseModel
• Implementation:Elevenlabs_Elevenlabs_python_AudioOutputMulti