This example transcribes an audio file that mixes english and japanese using Whisper in TXT, VTT and SRT formats.
Then uses DeepL API to translate the generated transcripts to spanish.
task: Transcribe
audio_file: english_japanese.mp3
use_model: large-v2
language: Auto-Detect (Japanese)
Detected language: Japanese
[00:00.000 --> 00:07.000] This is Unit 1 of Pimsleur's Japanese 1. Listen to this Japanese conversation.
[00:07.000 --> 00:10.000] すみません。英語がわかりますか?
[00:10.000 --> 00:15.000] いいえ、わかりません。日本語がわかりますか?
[00:15.000 --> 00:17.000] はい、少しわかります。
[00:17.000 --> 00:19.000] あなたはアメリカ人ですか?
[00:19.000 --> 00:34.000] はい、私はアメリカ人です。
output_formats: txt,vtt,srt
audio_transcription/english_japanese.txt
audio_transcription/english_japanese.vtt
audio_transcription/english_japanese.srt
deepl_target_language: Spanish
deepl_formality: default
deepl_coherence_preference: Share context between lines
DeepL: Translate results from Japanese [JA] to Spanish [ES]
[00:00.000 --> 00:07.000] Esta es la Unidad 1 de Japonés 1 de Pimsleur. Escuche esta conversación en japonés.
[00:07.000 --> 00:10.000] Disculpe. ¿Entiende inglés?
[00:10.000 --> 00:15.000] No, no entiendo. ¿Entiende japonés?
[00:15.000 --> 00:17.000] Sí, entiendo un poco.
[00:17.000 --> 00:19.000] ¿Es usted americano?
[00:19.000 --> 00:34.000] Sí, soy americano.
audio_transcription/english_japanese_Spanish.txt