You can use the AssemblyAI Audio Transcript Loader from LlamaHub to transcribe audio files inside your LlamaIndex applications.

Quickstart

First, install the assemblyai python package.

$
pip install assemblyai

Set your AssemblyAI API key as an environment variable named ASSEMBLYAI_API_KEY. You can get a free AssemblyAI API key from the AssemblyAI dashboard.

$
# Mac/Linux:
>
export ASSEMBLYAI_API_KEY=<YOUR_API_KEY>
>
>
# Windows:
>
set ASSEMBLYAI_API_KEY=<YOUR_API_KEY>

1. To load and transcribe audio data into documents,
2. Configure the file_path argument with a URL or a local file path to an audio or video file.

1
from llama_hub.assemblyai 
2
3
audio_file = "https://assembly.ai/nbc.mp3"
4
# or a local file path: audio_file = "./nbc.mp3"
5
6
reader = AssemblyAIAudioTranscriptReader(file_path=audio_file)
7
8
docs = reader.load_data()

The reader.load_data() method returns an array of documents, but by default, there’s only one document in the array with the full transcript. The transcribed text is available in the text attribute:

1
docs[0].text
2
# "Load time, a new president and new congressional makeup. Same old ..."

The metadata contains the full transcript object with more meta information:

1
docs[0].metadata
2
# {'language_code': <LanguageCode.en_us: 'en_us'>,
3
#  'audio_url': 'https://assembly.ai/nbc.mp3',
4
#  'punctuate': True,
5
#  'format_text': True,
6
#   ...
7
# }

Transcript formats

You can specify the transcript_format argument to load the transcript in different formats.

Depending on the format, load_data() returns either one or more documents. These are the different TranscriptFormat options:

• TEXT: One document with the transcription text
• SENTENCES: Multiple documents, splits the transcription by each sentence
• PARAGRAPHS: Multiple documents, splits the transcription by each paragraph
• SUBTITLES_SRT: One document with the transcript exported in SRT subtitles format
• SUBTITLES_VTT: One document with the transcript exported in VTT subtitles format

1
from llama_hub.assemblyai 
2
3
reader = AssemblyAIAudioTranscripReader(
4
    file_path="./your_file.mp3",
5
    transcript_format=TranscriptFormat.SENTENCES,
6
)
7
8
docs = reader.load_data()

Transcription config

You can also specify the config argument to use different audio intelligence models.

1
import assemblyai as aai
2
3
config = aai.TranscriptionConfig(speaker_labels=True,
4
                                 auto_chapters=True,
5
                                 entity_detection=True
6
)
7
8
reader = AssemblyAIAudioTranscriptReader(
9
    file_path="./your_file.mp3",
10
    config=config
11
)

Pass the API key as argument

You can also pass the AssemblyAI API key as an argument instead of an environment variable.

1
reader = AssemblyAIAudioTranscriptReader(
2
    file_path="./your_file.mp3",
3
    api_key="<YOUR_API_KEY>"
4
)

Additional resources

You can learn more about using LlamaIndex with AssemblyAI in these resources.

• How to use audio data in LlamaIndex with Python
• Build LlamaIndex Audio Apps with Python in 5 minutes
• LlamaHub AssemblyAI documentation
• LlamaHub AssemblyAI source code
• AssemblyAI Python SDK