I transscribed a German video. Unfortunaltey, the transscriptioin was in English, and I had to translate it back to German again. Of course, in this process much of the original content was lost ("reading between the lines"). Is it possible to transscribe in the original language, without the deviation via English? How?
Absolutely, before you hit the transcribe button, there is an option to select the language of the input audio. I highly recommend always setting this up correctly. Sometimes the AI can pick up and do translations on the fly but this is not very reliable.
Hence, when you click "New transcription", under settings there, select German and then retranscribe the audio. This will 10x the...
I record client meetings through Apple Voice Memo where it is two speakers on one audio input. Most transcription software can easily ID two speakers when you use Zoom, Teams, etc because you have an input from each user account in the meeting. Can this identify two speakers by “Speaker 1” or “Speaker 2” when both are coming in through one account? I currently use Otter.AI which can easily do this, but the transcripts tend to have a lot of errors and have a hard time when one speaker interrupts the other, or it drops whole sentences when the speakers talk too close together. I tried creating a Make.com scenario, but it won’t ID each speaker and fails if the recording is over 45 minutes. My client meetings can last up to 120 - 150 minutes in one recording.
yes it can identify multiple speakers (up to 10) automatically. The only limitation currently is that we only offer the speaker recognition model for the minutes that come with your plan and not the BYOK part. The reason for this is that we currently only let users add an OpenAI API key. Unfortunately OpenAI doesn't have speaker recognition in their own model deployment.
A: The transcription quality is very different for those two. Additionally: 1. We provide world level timestamps and subtitle formats to add to a video 2. We let you automatically create the same streamlined content every time for each transcript. Even if you are pretty experienced in ChatGPT, this saves many of our users 1-2 hours per week 3. You can create video clips in our app 4. You can...
that shouldn't be a problem at all! We support uploading the following formats: 'mp3', 'mp4', 'm4a', 'mpeg', 'mpga', 'wav', 'webm', 'ogg', 'flac', 'aac', 'opus'
Keep in mind though that we don't have a phone app so from the phone you still need to get it to your computer.
We will in the future also have a phone app but that is still under development.
For this to be useful for me I would need to be able to use this tool by way of automations. Is there API access, or a way to get transcriptions via http request, or other means of automation in n8n?
A: No, that is currently not there! We might have some workflow automatons later on but if you just need a transcript, I would recommend automating your workflow directly via the OpenAI API and n8n.
Q: Transscription in original language?
I transscribed a German video. Unfortunaltey, the transscriptioin was in English, and I had to translate it back to German again. Of course, in this process much of the original content was lost ("reading between the lines"). Is it possible to transscribe in the original language, without the deviation via English? How?
Laurin_WhisperTranscribe
Apr 24, 2025A: Hi!
Absolutely, before you hit the transcribe button, there is an option to select the language of the input audio. I highly recommend always setting this up correctly. Sometimes the AI can pick up and do translations on the fly but this is not very reliable.
Hence, when you click "New transcription", under settings there, select German and then retranscribe the audio. This will 10x the...
Share WhisperTranscribe
Verified purchaser
Yes, that worked perfectly! 👍 Thank you.
Q: Two person identification
I record client meetings through Apple Voice Memo where it is two speakers on one audio input. Most transcription software can easily ID two speakers when you use Zoom, Teams, etc because you have an input from each user account in the meeting. Can this identify two speakers by “Speaker 1” or “Speaker 2” when both are coming in through one account? I currently use Otter.AI which can easily do this, but the transcripts tend to have a lot of errors and have a hard time when one speaker interrupts the other, or it drops whole sentences when the speakers talk too close together. I tried creating a Make.com scenario, but it won’t ID each speaker and fails if the recording is over 45 minutes. My client meetings can last up to 120 - 150 minutes in one recording.
Laurin_WhisperTranscribe
Apr 24, 2025A: Hi there,
yes it can identify multiple speakers (up to 10) automatically. The only limitation currently is that we only offer the speaker recognition model for the minutes that come with your plan and not the BYOK part. The reason for this is that we currently only let users add an OpenAI API key. Unfortunately OpenAI doesn't have speaker recognition in their own model deployment.
We offer...
Share WhisperTranscribe
Q: What is the difference between your product and transcribing from Word and then uploading to ChatGPT?
Laurin_WhisperTranscribe
Apr 24, 2025A: The transcription quality is very different for those two. Additionally:
1. We provide world level timestamps and subtitle formats to add to a video
2. We let you automatically create the same streamlined content every time for each transcript. Even if you are pretty experienced in ChatGPT, this saves many of our users 1-2 hours per week
3. You can create video clips in our app
4. You can...
Share WhisperTranscribe
Q: Can I upload a memo from a voicerecorder?
I would like to generate transcripts from the voicerecorder from my Iphone. Can I upload those memo's to Whispertranscribe?
Laurin_WhisperTranscribe
Apr 23, 2025A: Hi Leintje,
that shouldn't be a problem at all! We support uploading the following formats:
'mp3', 'mp4', 'm4a', 'mpeg', 'mpga', 'wav', 'webm', 'ogg', 'flac', 'aac', 'opus'
Keep in mind though that we don't have a phone app so from the phone you still need to get it to your computer.
We will in the future also have a phone app but that is still under development.
Share WhisperTranscribe
Q: API Access?
For this to be useful for me I would need to be able to use this tool by way of automations. Is there API access, or a way to get transcriptions via http request, or other means of automation in n8n?
Laurin_WhisperTranscribe
Apr 23, 2025A: No, that is currently not there! We might have some workflow automatons later on but if you just need a transcript, I would recommend automating your workflow directly via the OpenAI API and n8n.
Share WhisperTranscribe