1 ratio. However, this can vary depending on the factors mentioned above.
For example, if the audio is of poor quality or has background noise, it may take longer to transcribe. Additionally, if the speaker speaks quickly or has a strong accent, it may also take longer to transcribe.
Also worth reading: Can Offline Embedded Voice AI Tools Really Transcribe Audio Without the Cloud? · How do Gemini 3.5, MAI-Transcribe, and Qwen-Audio stack up in AI transcription accuracy benchmarks for 2026? · What makes on-device speech to text the most private way to transcribe audio in 2025?
On the other hand, some transcriptionists may be able to transcribe audio at a faster rate, such as 30 minutes of audio in an hour. However, this is generally not the industry standard and may not be the case for most transcriptionists.
In summary, the amount of time it takes to transcribe 10 minutes of audio can vary depending on several factors, but on average, it would take around 2-4 hours to transcribe one hour of clear audio.